2025-11-21_de-ae-ai_standup_e6dc5951

Meeting Title: DE-AE-AI Standup Date: 2025-11-21 Meeting participants: Awaish Kumar, Mustafa Raja, Gabriel Lam, Uttam Kumaran, Casie Aviles, Samuel Roberts, Rico Rejoso, Ashwini Sharma, Zoran Selinger, Amber Lin, Robert Tseng

---

WEBVTT

1 00:01:18.170 ⇒ 00:01:19.610 Uttam Kumaran: Hey guys, good morning.

2 00:01:19.800 ⇒ 00:01:20.930 Uttam Kumaran: Good evening.

3 00:01:21.360 ⇒ 00:01:22.430 Mustafa Raja: Ayy.

4 00:01:22.780 ⇒ 00:01:23.550 Awaish Kumar: Hello.

5 00:01:25.460 ⇒ 00:01:26.540 Awaish Kumar: Good evening.

6 00:01:28.060 ⇒ 00:01:29.019 Uttam Kumaran: How’s it going?

7 00:01:31.170 ⇒ 00:01:31.590 Mustafa Raja: Good.

8 00:01:31.590 ⇒ 00:01:32.260 Awaish Kumar: I’m good.

9 00:01:35.510 ⇒ 00:01:41.060 Uttam Kumaran: Yeah. It’s, it’s Thanksgiving week here next week, guys, so…

10 00:01:41.340 ⇒ 00:01:46.549 Uttam Kumaran: It’s, like, big holidays here in the States, but I don’t… I don’t think anyone else in the world celebrates.

11 00:01:46.610 ⇒ 00:01:47.200 Gabriel Lam: Heh, no.

12 00:01:47.200 ⇒ 00:01:48.519 Samuel Roberts: No, I don’t think so.

13 00:01:50.550 ⇒ 00:02:04.020 Samuel Roberts: Yeah, so I will be off Thursday, Friday next week, pretty much out of contact, probably, and then I’ll be traveling the next week, so I will probably be mostly out of contact, but I’ll be able to be on Slack probably the next week.

14 00:02:05.810 ⇒ 00:02:13.739 Uttam Kumaran: Yeah, for the most part, I think all the U.S. folks… I drafted a note, I’ll be sending it. All the US folks will mostly be off.

15 00:02:13.930 ⇒ 00:02:18.750 Uttam Kumaran: Thursday, Friday, unless they choose otherwise.

16 00:02:18.920 ⇒ 00:02:21.060 Uttam Kumaran: But it is a day off, and then…

17 00:02:21.320 ⇒ 00:02:24.739 Uttam Kumaran: Also in… for Christmas, we’ll… we’ll…

18 00:02:24.900 ⇒ 00:02:31.960 Uttam Kumaran: earmarks and time off. None of our clients will be on. Most of them, in fact, will… a lot of them are off the entirety of next week.

19 00:02:32.190 ⇒ 00:02:33.910 Uttam Kumaran: So it will be a slow week.

20 00:02:33.910 ⇒ 00:02:34.580 Samuel Roberts: Oh, wow.

21 00:02:34.810 ⇒ 00:02:39.560 Uttam Kumaran: Yeah, like, some… I think Urban Stems, a lot of them are off.

22 00:02:42.090 ⇒ 00:02:50.219 Uttam Kumaran: Yeah, and a few folks, our core stakeholders, so in terms of next week, we’ll probably look to, like, ship a couple things by Wednesday, and then sort of call it.

23 00:02:50.460 ⇒ 00:02:54.930 Uttam Kumaran: My family’s also visiting me here, so I’ll be…

24 00:02:54.930 ⇒ 00:02:55.820 Samuel Roberts: Wellness.

25 00:02:56.180 ⇒ 00:03:07.599 Uttam Kumaran: I’ll be in and out. Yeah, they’re coming from California, we’re gonna do Thanksgiving here this year. Oh, cool. Yeah, I’ll be going to my wife’s family for Thanksgiving, so maybe I’ll be online if I need to escape at some point. We’ll see.

26 00:03:11.340 ⇒ 00:03:16.020 Uttam Kumaran: Yeah, I mean, I’m gonna be… I can only think about work. What else is there to think about, so…

27 00:03:16.020 ⇒ 00:03:17.330 Samuel Roberts: Right, right.

28 00:03:17.330 ⇒ 00:03:25.360 Uttam Kumaran: I’ll be thinking about work, but… yeah, and then we have, Element will be starting on December 1st.

29 00:03:26.880 ⇒ 00:03:33.019 Uttam Kumaran: So, that one, I think we’ll… we can talk a little bit Monday, but most likely, it will start,

30 00:03:33.490 ⇒ 00:03:35.349 Uttam Kumaran: with Awash, you, and me.

31 00:03:36.350 ⇒ 00:03:40.550 Awaish Kumar: Are there any, like, notes from sales, like.

32 00:03:40.550 ⇒ 00:03:43.459 Uttam Kumaran: Yes, there’s a lot… I have a lot for you, yeah.

33 00:03:45.350 ⇒ 00:03:48.019 Awaish Kumar: Can I log into HubSpot and have access to those?

34 00:03:50.750 ⇒ 00:03:58.399 Uttam Kumaran: Yeah, I don’t know if they’re… because all the notes are in both the doc, Gantt chart, and then in the, platform.

35 00:03:58.850 ⇒ 00:04:05.529 Uttam Kumaran: So if you search for Element in the platform, you’ll see the meetings, and then I’ll add you… maybe, Rico, we can… let’s include Awash in the channel.

36 00:04:05.800 ⇒ 00:04:15.599 Uttam Kumaran: And then if you can shoot him all of the materials, he can check it out. This is like an Urban Stems. Think about Element as, like, Urban Stems, except if we came in with nothing.

37 00:04:18.360 ⇒ 00:04:19.080 Uttam Kumaran: So, I…

38 00:04:19.089 ⇒ 00:04:22.290 Awaish Kumar: does not have a data infrastructure, right?

39 00:04:22.400 ⇒ 00:04:23.550 Uttam Kumaran: No. Okay.

40 00:04:24.800 ⇒ 00:04:26.209 Uttam Kumaran: Yeah, it didn’t have it.

41 00:04:37.370 ⇒ 00:04:38.050 Uttam Kumaran: Cool.

42 00:04:38.720 ⇒ 00:04:43.990 Uttam Kumaran: Yeah, I guess let’s… maybe we can start with… maybe we start with Urban Stems Awash? Like, I just sent a note.

43 00:04:44.780 ⇒ 00:04:50.630 Awaish Kumar: Yeah, I’ll… So, for urban stems, like, for the snapshot model itself.

44 00:04:50.720 ⇒ 00:05:02.329 Awaish Kumar: The models are, like, working fine. I don’t see any issues with snapshot models themselves. Like, they’re just snapshots themselves, but the underlying model

45 00:05:02.330 ⇒ 00:05:11.710 Awaish Kumar: which we are generating, they are pretty big, because we then have to generate, like, hourly, kind of a rope,

46 00:05:12.520 ⇒ 00:05:14.490 Awaish Kumar: For… for all the…

47 00:05:14.930 ⇒ 00:05:23.850 Awaish Kumar: Inventory details, and that makes it, like, a huge table, and it takes, like, hours to finish, if we have to run full refreshes.

48 00:05:23.990 ⇒ 00:05:27.770 Awaish Kumar: So I don’t think there’s any,

49 00:05:28.040 ⇒ 00:05:39.260 Awaish Kumar: like, the requirement, like, I can look into optimizations, but, like, there’s more, like, a need for a process. So once we change any schema.

50 00:05:39.520 ⇒ 00:05:47.700 Awaish Kumar: of a model which is, like, really huge, how we are going to handle those schema changes? Like, that’s the question.

51 00:05:49.880 ⇒ 00:05:55.559 Uttam Kumaran: I mean, why don’t we instead snapshot, like, the underlying tables? They’re smaller.

52 00:05:59.240 ⇒ 00:06:05.960 Awaish Kumar: Yeah, snapshots themselves are not that big, but then we… we are generating a model out of snapshots.

53 00:06:06.080 ⇒ 00:06:11.260 Awaish Kumar: And that’s… that gives us the hourly view, and that’s really big itself.

54 00:06:15.000 ⇒ 00:06:15.880 Uttam Kumaran: I see.

55 00:06:16.630 ⇒ 00:06:23.300 Uttam Kumaran: So one… I mean, one question you should ask is, do they need the hourly view for all time?

56 00:06:25.950 ⇒ 00:06:27.910 Uttam Kumaran: Or do they just need, like, last…

57 00:06:28.320 ⇒ 00:06:29.210 Awaish Kumar: Yeah, yeah, yeah.

58 00:06:29.210 ⇒ 00:06:30.610 Uttam Kumaran: Month, last month.

59 00:06:30.970 ⇒ 00:06:32.440 Uttam Kumaran: Maybe ask them that.

60 00:06:32.580 ⇒ 00:06:37.920 Uttam Kumaran: Because I don’t know if they’re using historical hourly, and they can always rebuild it if they need, when they need it, you know?

61 00:06:38.070 ⇒ 00:06:51.390 Awaish Kumar: Yeah, I can ask for it, I can look into the queries if I can optimize anything, but otherwise, we just need a process once, if we make any, like, we are running those models incrementally.

62 00:06:52.040 ⇒ 00:06:59.849 Awaish Kumar: Right? So, also, like, we are running them incrementally, there’s no problem in,

63 00:07:00.120 ⇒ 00:07:08.079 Awaish Kumar: recreating those models if they just run incrementally. The problem arrives when there is a schema change.

64 00:07:08.480 ⇒ 00:07:15.909 Awaish Kumar: And we need full refreshes. So, the only way to handle that is that we come up with a process that

65 00:07:15.940 ⇒ 00:07:30.609 Awaish Kumar: Whenever we have this schema changes, we are going to run it… still, we are going to run it incrementally, but we just need to know what is the timeframe for which we need to run it. Like, maybe last 3 months are required.

66 00:07:30.610 ⇒ 00:07:39.300 Awaish Kumar: for this new field, and we don’t fill, like, rerun, like, entirely. We just rerun last 3 months of data, or whatever.

67 00:07:39.710 ⇒ 00:07:46.330 Uttam Kumaran: Maybe… maybe that’s, like, an SOP we create, because ultimately, it’s gonna be up to Emily to run that, you know?

68 00:07:47.050 ⇒ 00:07:47.740 Awaish Kumar: -

69 00:07:49.850 ⇒ 00:07:51.809 Awaish Kumar: Okay, I understand.

70 00:07:51.810 ⇒ 00:07:52.430 Uttam Kumaran: Yeah.

71 00:07:52.430 ⇒ 00:07:52.870 Awaish Kumar: Of course.

72 00:07:52.870 ⇒ 00:07:56.929 Uttam Kumaran: Ask them… ask them about the hourly table, like, if they truly need…

73 00:07:57.900 ⇒ 00:08:04.060 Uttam Kumaran: like, if they truly need that much data. Second thing is to ask about, yeah.

74 00:08:04.060 ⇒ 00:08:15.740 Awaish Kumar: like, Demilade said that she needs, like, there was an issue in revenue data some time ago, and I asked him, like, do we really need historical data? And he said, like, Emily is…

75 00:08:15.900 ⇒ 00:08:19.469 Awaish Kumar: Using historical data for her analysis.

76 00:08:20.330 ⇒ 00:08:20.990 Uttam Kumaran: Okay.

77 00:08:25.810 ⇒ 00:08:27.899 Uttam Kumaran: But historically, hourly data?

78 00:08:29.790 ⇒ 00:08:33.450 Awaish Kumar: Yes, like, shim… yeah, that doesn’t make…

79 00:08:33.650 ⇒ 00:08:40.440 Awaish Kumar: sense, right? But I will confirm with Emily. Maybe, you know, I will try to raise this kind of…

80 00:08:40.760 ⇒ 00:08:42.140 Awaish Kumar: To ask her.

81 00:08:42.620 ⇒ 00:08:43.600 Uttam Kumaran: Okay, okay.

82 00:08:44.920 ⇒ 00:08:46.140 Uttam Kumaran: Okay, great.

83 00:08:46.780 ⇒ 00:08:51.509 Uttam Kumaran: And then I’m… yeah, I’m still gonna work on full refresh fixes, so it’s fine.

84 00:08:51.970 ⇒ 00:08:54.730 Uttam Kumaran: And then next week is our…

85 00:08:54.880 ⇒ 00:09:00.020 Uttam Kumaran: We’re coming up on renewal. Basically, what’s gonna happen is…

86 00:09:00.130 ⇒ 00:09:08.360 Uttam Kumaran: Zach… I mean, it’s… I can tell you about it later, but basically, Zach wants to slim down our support, because…

87 00:09:08.810 ⇒ 00:09:12.850 Uttam Kumaran: Right now, he doesn’t underst- nope, it’s not really clear, like…

88 00:09:12.960 ⇒ 00:09:22.040 Uttam Kumaran: whether the team can sustain itself or not, and I think he wants to make some changes. So we’re gonna kind of move to more of a support retainer.

89 00:09:22.720 ⇒ 00:09:25.429 Uttam Kumaran: And then, most likely come back.

90 00:09:26.330 ⇒ 00:09:29.849 Uttam Kumaran: you know, in… in February, or in…

91 00:09:30.450 ⇒ 00:09:35.049 Uttam Kumaran: Yeah, probably in January or February, and come back with a larger scope.

92 00:09:36.860 ⇒ 00:09:46.079 Uttam Kumaran: So, kind of like, this is going to be just us closing things out this week and next week, so… at least until… so that we just move into more of a support mode.

93 00:09:47.850 ⇒ 00:09:48.510 Awaish Kumar: Okay.

94 00:09:49.740 ⇒ 00:09:55.029 Uttam Kumaran: Which is good, because I think we’ll want to take some of your time for Element anyway, so…

95 00:09:57.500 ⇒ 00:10:03.859 Awaish Kumar: Yeah, that would be… I think… I’m excited for Element. They don’t have infrastructure. We’re going to lay down.

96 00:10:04.790 ⇒ 00:10:06.770 Uttam Kumaran: Same, yeah, so it’ll… it’ll be…

97 00:10:07.060 ⇒ 00:10:11.429 Uttam Kumaran: Me and you to start, and then we’ll loop other people in as…

98 00:10:11.930 ⇒ 00:10:14.630 Uttam Kumaran: As we go there, so… okay.

99 00:10:15.330 ⇒ 00:10:20.159 Uttam Kumaran: Great, so that’s Urban Stems. Anything, I guess…

100 00:10:20.470 ⇒ 00:10:25.149 Uttam Kumaran: Anything on the data engine or AE side for Eden?

101 00:10:26.480 ⇒ 00:10:36.359 Awaish Kumar: Yeah, I shipped a model just, like, yesterday. Basically, that’s about, like this…

102 00:10:36.500 ⇒ 00:10:46.980 Awaish Kumar: KP model, like, for the marketing team, for meta KP API, they needed an attribution table, and I have worked on that model

103 00:10:47.190 ⇒ 00:10:55.150 Awaish Kumar: basically combining our… their backend data, which is coming from Bask, with the age layer, which is a run build, and some…

104 00:10:55.370 ⇒ 00:10:58.649 Awaish Kumar: Intake flow data, and then…

105 00:10:59.350 ⇒ 00:11:09.830 Awaish Kumar: I’ve generated a table, which will be used to basically reverse ETL the data to meta for

106 00:11:10.070 ⇒ 00:11:20.389 Awaish Kumar: to identify the, like, the orders which they need to pay for, and the model is ready. I… then, I just need Zoran’s…

107 00:11:20.590 ⇒ 00:11:34.389 Awaish Kumar: Zoran to just look at it and validate it, and after that, we have a task for basically reverse detailing that. I’m not sure if Zoran is going to handle that, or he will send it over to me.

108 00:11:37.710 ⇒ 00:11:40.390 Uttam Kumaran: Okay, do you want to ping him in the channel?

109 00:11:40.640 ⇒ 00:11:42.190 Awaish Kumar: To let them know.

110 00:11:42.300 ⇒ 00:11:43.879 Uttam Kumaran: And then we can discuss that in next year.

111 00:11:45.060 ⇒ 00:11:47.459 Awaish Kumar: Yeah, yeah, I, I can… I can be here.

112 00:11:48.010 ⇒ 00:11:48.810 Uttam Kumaran: Okay.

113 00:11:49.210 ⇒ 00:11:52.760 Uttam Kumaran: Okay, cool. And then, I know Ashwini handled one couple things.

114 00:11:53.230 ⇒ 00:11:54.000 Uttam Kumaran: Are you gonna hand off?

115 00:11:54.410 ⇒ 00:11:55.399 Uttam Kumaran: Anything else to him?

116 00:11:56.250 ⇒ 00:12:01.610 Awaish Kumar: So far, Eden, I don’t think so, for this week.

117 00:12:03.940 ⇒ 00:12:04.640 Uttam Kumaran: Okay.

118 00:12:05.300 ⇒ 00:12:08.520 Awaish Kumar: Yeah, there’s no AED requirement this week.

119 00:12:09.050 ⇒ 00:12:12.030 Awaish Kumar: So he handled the catalyst thing, which…

120 00:12:12.480 ⇒ 00:12:17.370 Awaish Kumar: Was it still kind of incoming requests, like, it was…

121 00:12:17.480 ⇒ 00:12:20.010 Awaish Kumar: But for now, that just works as it is.

122 00:12:20.620 ⇒ 00:12:21.260 Uttam Kumaran: Okay.

123 00:12:22.080 ⇒ 00:12:31.269 Uttam Kumaran: So, I think it would be great if, yeah, like, we can plan out today when a little bit of more, like, tickets for him next week. I want him to handle the metaplane thing.

124 00:12:32.400 ⇒ 00:12:39.530 Uttam Kumaran: And then… And then start to take on additional scope next week.

125 00:12:39.970 ⇒ 00:12:44.709 Uttam Kumaran: And then ideally, It’s sort of… you can kind of, like, just mainly…

126 00:12:45.200 ⇒ 00:12:50.610 Uttam Kumaran: Assist them and sort of hand off and assign, and then you can come with me on to the next clients.

127 00:12:51.600 ⇒ 00:12:53.370 Uttam Kumaran: Okay. That’d be great.

128 00:12:55.270 ⇒ 00:13:08.850 Uttam Kumaran: And then for Ashrini as well, I think maybe, Awish, we can decide on Monday where everybody is, but I’m having Sam help me with CCA, which is great, because I actually think this client is gonna be…

129 00:13:09.390 ⇒ 00:13:15.200 Uttam Kumaran: It’s not gonna be so fast-paced and technical, it’s actually gonna be…

130 00:13:15.510 ⇒ 00:13:21.059 Uttam Kumaran: The difficulty’s gonna come from working with, like, a really large org.

131 00:13:21.380 ⇒ 00:13:25.420 Uttam Kumaran: And so, I actually prefer to have someone

132 00:13:25.670 ⇒ 00:13:31.229 Uttam Kumaran: here, because I don’t know when meetings are gonna be, and, like, it’s gonna be a little bit all over, so it’s gonna be a lot of FaceTime.

133 00:13:31.380 ⇒ 00:13:36.459 Uttam Kumaran: So, I think me and Sam will handle to start, and then we will loop in…

134 00:13:36.670 ⇒ 00:13:42.180 Uttam Kumaran: Most likely loop in Ashwini to execute a lot of the things.

135 00:13:42.490 ⇒ 00:13:54.479 Uttam Kumaran: So, like, it’ll be me, kind of, on the strategy side. Sam, I’ll have you do, like, architecture, larger planning, and, like, kind of, like, big infrastructure execution, and then as soon as things start to, like, stabilize, and we have, like.

136 00:13:54.890 ⇒ 00:14:02.249 Uttam Kumaran: two to three weeks’ worth of roadmap, we can loop in Ashwini, because we’ll get into… we’ll get into a lot of, like, modeling and things like that, and…

137 00:14:02.400 ⇒ 00:14:07.800 Uttam Kumaran: Yeah. And given the amount of sources, we’ll need to sort of… Triple T, so…

138 00:14:10.690 ⇒ 00:14:17.110 Uttam Kumaran: Cool, okay, so that… and then for, yeah, if we transition to…

139 00:14:17.630 ⇒ 00:14:22.509 Uttam Kumaran: CTA stuff, yeah, I feel good. I talked to Catherine again yesterday.

140 00:14:22.510 ⇒ 00:14:23.150 Samuel Roberts: Okay.

141 00:14:23.940 ⇒ 00:14:41.120 Uttam Kumaran: Yeah, I mean, we’re… our pace is really, really good. I mean, it’s… I think they’re… they’re happy, so… I… I want to put in front of her today, like, just drafts of the diagram and the Gantt chart, so I think, Sam, thanks for just even taking a crack at that. Like, I just saw…

142 00:14:41.120 ⇒ 00:14:45.429 Samuel Roberts: I wasn’t really sure what was most important to include in some of those things yet, because I’m still…

143 00:14:45.430 ⇒ 00:14:52.759 Uttam Kumaran: I mean, it’s just helpful even to just get something on paper, and then it’ll be something we work with them on, so we’re not, like…

144 00:14:53.020 ⇒ 00:15:10.100 Uttam Kumaran: We’re not, prescribing the whole thing, but we… we will own, sort of, like, the organization of it all, so… Yeah. So that’s something I’ll get to her, and also send her over the diagram, which I’ll kind of poke at today, so… I feel pretty good. I guess my…

145 00:15:10.790 ⇒ 00:15:13.410 Uttam Kumaran: If anything, for today, like…

146 00:15:13.780 ⇒ 00:15:20.500 Uttam Kumaran: I think the… one thing I also owe, and I didn’t get a chance to do, is sort of, like, what Snowflake setup is gonna look like. Yes.

147 00:15:21.030 ⇒ 00:15:28.640 Uttam Kumaran: So we can either, we can either do that today together, or…

148 00:15:29.020 ⇒ 00:15:31.310 Uttam Kumaran: I don’t know, kind of decide about it?

149 00:15:31.460 ⇒ 00:15:34.470 Uttam Kumaran: So I’m trying to think about…

150 00:15:35.210 ⇒ 00:15:37.270 Uttam Kumaran: I mean, we could also spend the…

151 00:15:38.320 ⇒ 00:15:40.970 Uttam Kumaran: Back half of this meeting, working on that together.

152 00:15:41.880 ⇒ 00:15:47.700 Uttam Kumaran: like, me and Mustafa can work on, like, the dbt initialization script, and…

153 00:15:48.060 ⇒ 00:15:50.599 Uttam Kumaran: Snowflake initialization, if you think that’s good.

154 00:15:51.770 ⇒ 00:15:54.579 Samuel Roberts: Yeah, I mean, if that’s the time you’ve got, definitely.

155 00:15:54.960 ⇒ 00:15:58.760 Uttam Kumaran: Cool, okay, okay, okay, great. And let’s do that.

156 00:15:59.320 ⇒ 00:16:02.820 Uttam Kumaran: Okay. Cool. So, I feel good about,

157 00:16:03.270 ⇒ 00:16:09.919 Uttam Kumaran: CTA? Yeah, probably the only thing, Sam, is, like, they’re gonna have meetings randomly, so I may…

158 00:16:10.120 ⇒ 00:16:24.590 Uttam Kumaran: like, I’m just gonna kind of just impart as much knowledge onto you, but I may have you kind of go attend some of those. The nice thing about CTA is they’re not very non-confrontational. As you can tell, even my, like, small amount of energy was very, like.

159 00:16:24.590 ⇒ 00:16:25.090 Samuel Roberts: Yeah.

160 00:16:25.090 ⇒ 00:16:33.769 Uttam Kumaran: impressive for them, so I wouldn’t… I wouldn’t worry about, like, understanding everything. I think more of the part is actually…

161 00:16:33.900 ⇒ 00:16:38.259 Uttam Kumaran: I kind of want you to start to see what it’s like to actually, like.

162 00:16:38.400 ⇒ 00:16:44.429 Uttam Kumaran: Go directly with the client, and then distill what they’re asking, and then, like, kind of see the path towards execution.

163 00:16:44.940 ⇒ 00:16:51.270 Uttam Kumaran: Right. So I think that will be amazing. That way, I’m… you’re able to kind of sit at that level.

164 00:16:51.370 ⇒ 00:16:54.069 Uttam Kumaran: Because that level is… that level… there’s…

165 00:16:54.340 ⇒ 00:17:01.089 Uttam Kumaran: And then for me, the biggest thing is, like, okay, I want to look at the entire project, and then also start to secure a scope for the future.

166 00:17:01.990 ⇒ 00:17:02.879 Samuel Roberts: Yeah, that makes sense.

167 00:17:02.880 ⇒ 00:17:06.440 Uttam Kumaran: So you kind of coming at that level really allows me to go do that, so…

168 00:17:07.119 ⇒ 00:17:17.080 Uttam Kumaran: I think that that will be, the plan there. So, and then, like, I’m on Slack, so even if, like, I’m not able to attend a meeting, if you Slack me, I can walk you through whatever.

169 00:17:17.150 ⇒ 00:17:36.639 Uttam Kumaran: And this… the initial part of this is gonna… is really gonna be sort of, like, pretty technical in terms of, like, Snowflake setup, and then it’s gonna get into a different kind of technical, which is more about, like, you know, each source. But it’s a lot of CRM and people data, so it’s not, like, so complicated.

170 00:17:36.640 ⇒ 00:17:37.480 Samuel Roberts: Right.

171 00:17:37.810 ⇒ 00:17:41.870 Uttam Kumaran: It’s not, like, sales or, like, really complicated e-commerce data, you know, so…

172 00:17:41.870 ⇒ 00:17:42.940 Samuel Roberts: Sure, sure.

173 00:17:43.360 ⇒ 00:17:44.110 Uttam Kumaran: Yeah.

174 00:17:44.580 ⇒ 00:17:45.920 Uttam Kumaran: Okay.

175 00:17:47.530 ⇒ 00:17:49.290 Uttam Kumaran: Okay. Cool.

176 00:17:50.580 ⇒ 00:17:52.470 Uttam Kumaran: Great.

177 00:17:53.180 ⇒ 00:18:01.500 Uttam Kumaran: And then, let’s talk about Hedra. So, yeah, I had a good meeting with Sandra yesterday evening, OH?

178 00:18:01.890 ⇒ 00:18:04.050 Uttam Kumaran: I think it would be great if…

179 00:18:04.530 ⇒ 00:18:11.770 Uttam Kumaran: Yeah, so I… I patched a couple… I sent some notes in… for about, kind of, patches. I think,

180 00:18:12.190 ⇒ 00:18:17.800 Uttam Kumaran: I… that meet… the meeting with her, I can send you the granola, but basically,

181 00:18:18.100 ⇒ 00:18:19.829 Uttam Kumaran: We need to, like…

182 00:18:20.330 ⇒ 00:18:26.940 Uttam Kumaran: think about the Stripe schema a little bit differently, like, I think we need to create a,

183 00:18:27.540 ⇒ 00:18:34.049 Uttam Kumaran: fact transactions… Which is just gonna be subscription and one-time transactions.

184 00:18:34.280 ⇒ 00:18:41.290 Uttam Kumaran: And then we also need to create, individual tables for the one time.

185 00:18:41.410 ⇒ 00:18:48.269 Uttam Kumaran: And the, like, and the subscriptions, because you can’t…

186 00:18:48.400 ⇒ 00:18:53.370 Uttam Kumaran: You can’t do ARR and MRR from, the one time.

187 00:18:54.200 ⇒ 00:18:55.989 Uttam Kumaran: Because they’re non-recurring.

188 00:18:56.580 ⇒ 00:18:57.520 Uttam Kumaran: So…

189 00:18:57.520 ⇒ 00:18:58.090 Awaish Kumar: Good.

190 00:18:58.090 ⇒ 00:19:00.490 Uttam Kumaran: That was the biggest thing.

191 00:19:00.490 ⇒ 00:19:00.980 Awaish Kumar: Hmm.

192 00:19:00.980 ⇒ 00:19:03.019 Uttam Kumaran: It’s for us to figure that out.

193 00:19:03.700 ⇒ 00:19:06.360 Awaish Kumar: So, like, on their platform, people…

194 00:19:06.670 ⇒ 00:19:09.490 Awaish Kumar: Can just pay once, without subscription.

195 00:19:10.960 ⇒ 00:19:13.940 Uttam Kumaran: No, no, no, it’s like, you’re buying, like, packs.

196 00:19:14.180 ⇒ 00:19:20.979 Uttam Kumaran: So, it’s like… it’s like, you already subscribed, and then you’re like, oh, I’m out of credits, so I want to buy more credits.

197 00:19:21.640 ⇒ 00:19:22.270 Awaish Kumar: Okay.

198 00:19:23.340 ⇒ 00:19:24.170 Uttam Kumaran: Yes.

199 00:19:26.650 ⇒ 00:19:30.949 Awaish Kumar: Like, we want to have a separate table for that, like, where we have a…

200 00:19:31.180 ⇒ 00:19:34.729 Uttam Kumaran: Well, we want to have one combined table for all revenue.

201 00:19:35.220 ⇒ 00:19:39.170 Uttam Kumaran: But then we also want a subscriptions table.

202 00:19:39.400 ⇒ 00:19:42.310 Uttam Kumaran: which can be used for MRR, ARR.

203 00:19:42.940 ⇒ 00:19:43.800 Awaish Kumar: Okay.

204 00:19:44.480 ⇒ 00:19:48.669 Uttam Kumaran: Because the transactions don’t count for MRR.

205 00:19:49.900 ⇒ 00:19:56.639 Awaish Kumar: So, we do have an existing subscription table, we just want to exclude the one-time.

206 00:19:56.640 ⇒ 00:19:58.749 Uttam Kumaran: Not in there. They’re not in there.

207 00:20:00.470 ⇒ 00:20:04.269 Uttam Kumaran: But that’s… I guess that’s what I’m saying. It’s not technically really by design.

208 00:20:04.410 ⇒ 00:20:06.500 Uttam Kumaran: Because you’re,

209 00:20:07.160 ⇒ 00:20:15.059 Uttam Kumaran: The way you’re handling the join between invoice items and subscriptions, you’re excluding a lot of the one-time… you’re excluding the one-time purchases.

210 00:20:15.250 ⇒ 00:20:19.849 Uttam Kumaran: So we need another table. We need both, basically, a one table just for those.

211 00:20:19.950 ⇒ 00:20:23.129 Uttam Kumaran: And then we need a table for all transactions.

212 00:20:26.600 ⇒ 00:20:26.980 Awaish Kumar: Okay.

213 00:20:26.980 ⇒ 00:20:28.060 Uttam Kumaran: Does that make sense?

214 00:20:28.930 ⇒ 00:20:32.719 Uttam Kumaran: We can also work through it later, but…

215 00:20:32.720 ⇒ 00:20:39.910 Awaish Kumar: Yeah, I looked at the, like, invoice lines… line items table, which is…

216 00:20:40.270 ⇒ 00:20:43.080 Awaish Kumar: Invoice items, basically, table, which is creative.

217 00:20:43.420 ⇒ 00:20:53.809 Awaish Kumar: which was created, and that’s why we had these negative amounts. Basically, what was happening is that this invoice item table, which is coming from Stripe.

218 00:20:53.970 ⇒ 00:20:56.970 Awaish Kumar: Has missing, rows.

219 00:20:57.560 ⇒ 00:21:09.230 Awaish Kumar: For items, like, it should have… if an invoice have 3 items, one is, like, total amount, two for discounts, then we have… should have 3… those 3 items in our items… invoice items table.

220 00:21:09.350 ⇒ 00:21:15.289 Awaish Kumar: But the problem is that some of… Some of those,

221 00:21:15.730 ⇒ 00:21:29.410 Awaish Kumar: invoices, only have two items, like, which are negative, and the original item, which basically has some positive amount, like, the original, the price, that’s missing, and that’s why it ends up in the negative

222 00:21:29.420 ⇒ 00:21:39.889 Awaish Kumar: amount at the end. So, I have created a PR, if you can review, like, I try to generate those items, like, synthetically, right? So.

223 00:21:40.140 ⇒ 00:21:44.710 Awaish Kumar: So, like, for those invoices where we However…

224 00:21:45.240 ⇒ 00:21:57.420 Awaish Kumar: Items… some of the items amount negative, but the total amount in an invoice is positive, then that means there is a missing item, which is… which is basically

225 00:21:57.540 ⇒ 00:22:01.730 Awaish Kumar: giving the actual price, so I just create… generated an

226 00:22:02.420 ⇒ 00:22:08.559 Awaish Kumar: Item-level thing in our int models, so at the end, it can give us the correct results.

227 00:22:09.700 ⇒ 00:22:10.360 Uttam Kumaran: Okay.

228 00:22:13.020 ⇒ 00:22:16.789 Awaish Kumar: So, yeah, I can share the PR, so you can review it.

229 00:22:27.140 ⇒ 00:22:41.279 Uttam Kumaran: Okay, yeah, if you send that to me, I can be looking… I can look at it. But I may ask you to… to grab time with her today, because I’m kind of busy, and I want… I… I think the… the revenue piece is a bit confusing, so…

230 00:22:41.610 ⇒ 00:22:46.620 Uttam Kumaran: I can take a look at your PR, but I think there’s still going to be a couple more models that we want to work on.

231 00:22:47.780 ⇒ 00:22:51.939 Awaish Kumar: Yeah, yeah, so I’m going to look at what you were saying about one-time things, and…

232 00:22:51.940 ⇒ 00:22:52.480 Uttam Kumaran: Yeah, okay.

233 00:22:52.480 ⇒ 00:22:57.999 Awaish Kumar: I’ll directly ask them, like, Whatever, like, the…

234 00:22:58.370 ⇒ 00:23:06.440 Awaish Kumar: whatever my questions will be, and maybe, like, they are really nice in answering those, so I will try to get it from them.

235 00:23:06.960 ⇒ 00:23:07.640 Uttam Kumaran: Okay.

236 00:23:20.110 ⇒ 00:23:21.900 Uttam Kumaran: Okay,

237 00:23:22.680 ⇒ 00:23:31.170 Uttam Kumaran: Great. And then, for README, Mustafa, I think we can talk about it in the next meeting, maybe, but I sent you,

238 00:23:32.180 ⇒ 00:23:33.169 Uttam Kumaran: I sent you the…

239 00:23:33.170 ⇒ 00:23:33.720 Mustafa Raja: Yeah.

240 00:23:33.720 ⇒ 00:23:35.230 Uttam Kumaran: Yeah.

241 00:23:35.450 ⇒ 00:23:42.310 Mustafa Raja: Yeah, those are the dashboard, updates. I haven’t thoroughly looked into it, but, that’s fine.

242 00:23:42.460 ⇒ 00:23:47.840 Mustafa Raja: Okay. And I’ll first, draft a plan on what needs to be in that

243 00:23:48.190 ⇒ 00:23:51.309 Mustafa Raja: Deliverable, and then work on that.

244 00:23:52.010 ⇒ 00:23:52.700 Uttam Kumaran: Okay.

245 00:23:55.680 ⇒ 00:23:56.880 Uttam Kumaran: Okay, that’s fine.

246 00:23:59.670 ⇒ 00:24:01.920 Uttam Kumaran: Hold on one second, this dog is barking.

247 00:24:39.270 ⇒ 00:24:44.240 Uttam Kumaran: Okay, so… Let’s talk about,

248 00:24:45.700 ⇒ 00:24:51.640 Uttam Kumaran: So, okay, so that’s where we are with Hedra. Anything on insomnia?

249 00:24:51.950 ⇒ 00:24:55.920 Uttam Kumaran: I think the only point I was gonna make is potentially,

250 00:24:56.530 ⇒ 00:25:02.829 Uttam Kumaran: Casey, for Amber to start looping you in, and a little bit of her analysis work that she’s doing that’s starting to mature?

251 00:25:04.440 ⇒ 00:25:04.940 Casie Aviles: Okay.

252 00:25:04.940 ⇒ 00:25:10.180 Uttam Kumaran: So, maybe I can… I was gonna ask her to do that, if you’re okay, and you feel like you have bandwidth.

253 00:25:11.600 ⇒ 00:25:16.219 Casie Aviles: Yeah, I think I should be able to do that with insomnia.

254 00:25:16.220 ⇒ 00:25:20.840 Uttam Kumaran: It’s gotten more… it’s gotten more mature, so I feel like…

255 00:25:21.500 ⇒ 00:25:28.529 Uttam Kumaran: It would be helpful, you know, for her, because she’s starting to work on a few other clients, so she can kind of hand off a couple things.

256 00:25:29.440 ⇒ 00:25:29.820 Casie Aviles: Sure.

257 00:25:30.720 ⇒ 00:25:31.750 Uttam Kumaran: Okay, cool.

258 00:25:32.090 ⇒ 00:25:36.569 Uttam Kumaran: And then the other thing on Insomnia is I’m gonna make another push.

259 00:25:37.140 ⇒ 00:25:40.989 Uttam Kumaran: to Robert to try to get us to improve the infrastructure.

260 00:25:41.300 ⇒ 00:25:49.819 Uttam Kumaran: Because… so, we kind of put a pause on that for a little bit, but I know it’s still really annoying every day to do that, so…

261 00:25:50.090 ⇒ 00:25:52.920 Uttam Kumaran: Yeah, like.

262 00:25:52.920 ⇒ 00:25:53.290 Casie Aviles: Exactly.

263 00:25:53.290 ⇒ 00:25:53.830 Uttam Kumaran: Oh, bye.

264 00:25:54.490 ⇒ 00:25:59.699 Uttam Kumaran: So I’m gonna… I’m trying, but they’re not really, like, seeing the vision, so…

265 00:25:59.700 ⇒ 00:26:00.260 Samuel Roberts: Yeah.

266 00:26:00.260 ⇒ 00:26:01.579 Uttam Kumaran: That’s a little bit hard.

267 00:26:04.540 ⇒ 00:26:09.839 Uttam Kumaran: Okay, great, so nothing on insomnia.

268 00:26:10.050 ⇒ 00:26:15.349 Uttam Kumaran: What else? Readme, Honey Stinger is fine.

269 00:26:15.680 ⇒ 00:26:22.480 Uttam Kumaran: Yeah, is there any other… oh, and then we talked about default yesterday, right? So…

270 00:26:22.600 ⇒ 00:26:27.419 Uttam Kumaran: Yeah, I think the only thing for default we have,

271 00:26:27.420 ⇒ 00:26:28.180 Mustafa Raja: The SAT.

272 00:26:29.250 ⇒ 00:26:33.159 Uttam Kumaran: is the, yeah, as the SOPs, and then also the,

273 00:26:33.590 ⇒ 00:26:38.139 Uttam Kumaran: The kind of comparison across all the vendors?

274 00:26:38.340 ⇒ 00:26:42.590 Uttam Kumaran: Maybe Ricoh also, like, you can create some tickets from…

275 00:26:42.820 ⇒ 00:26:45.829 Uttam Kumaran: That meeting, like, as a follow-up, that would be great.

276 00:26:53.570 ⇒ 00:26:58.160 Uttam Kumaran: Any other client stuff to cover?

277 00:27:01.990 ⇒ 00:27:07.770 Casie Aviles: I think for ABC, there’s just a migration plan.

278 00:27:07.770 ⇒ 00:27:08.180 Samuel Roberts: Yeah.

279 00:27:08.440 ⇒ 00:27:10.859 Casie Aviles: That we’ll be working on, so…

280 00:27:11.060 ⇒ 00:27:14.200 Casie Aviles: Okay. I booked a working session with.

281 00:27:14.440 ⇒ 00:27:17.589 Uttam Kumaran: Mustafa and some can be there optionally, but…

282 00:27:18.170 ⇒ 00:27:23.969 Casie Aviles: We’re also going to, like, create documentation for some, so we can easily

283 00:27:24.300 ⇒ 00:27:27.699 Casie Aviles: Like, understand, like, what’s going on with the stuff we built.

284 00:27:28.070 ⇒ 00:27:28.700 Uttam Kumaran: Okay.

285 00:27:29.680 ⇒ 00:27:31.970 Samuel Roberts: Yeah, I think kind of a two-step ticket, like.

286 00:27:32.090 ⇒ 00:27:36.689 Samuel Roberts: Document what’s there, and then figure out how to move it, rather than just dive right in, so…

287 00:27:37.690 ⇒ 00:27:40.989 Samuel Roberts: I think that’s… having that documentation will be very helpful, so…

288 00:27:43.540 ⇒ 00:27:48.780 Casie Aviles: Okay. Yeah, we’ll send out Notion links as we work on it.

289 00:27:51.190 ⇒ 00:27:52.110 Uttam Kumaran: Okay.

290 00:27:52.110 ⇒ 00:27:52.510 Samuel Roberts: Cool, cool.

291 00:27:53.110 ⇒ 00:27:58.790 Uttam Kumaran: Yeah, I’m, I’m meeting with, I’m meeting with…

292 00:27:59.170 ⇒ 00:28:03.899 Uttam Kumaran: the ABC team at their office on Monday, so I’m gonna be pitching them on…

293 00:28:03.900 ⇒ 00:28:04.420 Samuel Roberts: Yes.

294 00:28:04.680 ⇒ 00:28:13.470 Uttam Kumaran: the next Andy contract, on the Discovery kind of thing we’re pitching them on, and then also this, like, little migration, so… Okay.

295 00:28:13.740 ⇒ 00:28:14.510 Uttam Kumaran: Great.

296 00:28:16.390 ⇒ 00:28:22.140 Uttam Kumaran: Cool, I feel like that’s kind of it for… Clients…

297 00:28:22.320 ⇒ 00:28:29.770 Uttam Kumaran: I guess maybe, Gabe, I can hand it to you if there’s, pieces on,

298 00:28:31.910 ⇒ 00:28:35.029 Uttam Kumaran: Pieces on… what’s it called? Internal?

299 00:28:35.260 ⇒ 00:28:35.830 Gabriel Lam: Yep.

300 00:28:35.830 ⇒ 00:28:37.010 Samuel Roberts: Yeah.

301 00:28:37.010 ⇒ 00:28:49.030 Gabriel Lam: So yeah, we had two main pushes so far, one to do with the summary generation at the end of every meeting. Let me… oh, sorry, let me…

302 00:28:49.720 ⇒ 00:28:50.490 Gabriel Lam: Shoot.

303 00:28:51.110 ⇒ 00:28:56.569 Gabriel Lam: It’s not sharing a screen, really want to do that.

304 00:28:56.810 ⇒ 00:29:02.490 Gabriel Lam: The ingestion of every meeting, that we would have a pretty clear summary, so if anyone looks up

305 00:29:03.790 ⇒ 00:29:11.140 Gabriel Lam: to get the meeting at the end of it. At the end of… or whenever it’s uploaded, you have a pretty clear idea. That’s split into work streams, split into clients.

306 00:29:11.700 ⇒ 00:29:16.070 Gabriel Lam: I think there’s some… Information hierarchy that we can…

307 00:29:16.400 ⇒ 00:29:20.410 Gabriel Lam: cleanup. But you see client, work stream and team split into…

308 00:29:20.730 ⇒ 00:29:25.030 Gabriel Lam: different, Brainforce teams, as well as client teams.

309 00:29:25.230 ⇒ 00:29:28.040 Gabriel Lam: And so this is just a way for people to really…

310 00:29:28.290 ⇒ 00:29:32.920 Gabriel Lam: have a better overview of every meeting. These get ingested into…

311 00:29:33.090 ⇒ 00:29:38.239 Gabriel Lam: Linear tickets, so shoutout to Mustafa for getting this out, for persistence.

312 00:29:38.400 ⇒ 00:29:43.980 Gabriel Lam: We are hoping to increase the number of attributes that you can change here.

313 00:29:44.460 ⇒ 00:29:44.820 Uttam Kumaran: Great.

314 00:29:44.820 ⇒ 00:29:46.390 Gabriel Lam: See here, the body is…

315 00:29:46.970 ⇒ 00:29:47.510 Uttam Kumaran: A little bit.

316 00:29:47.510 ⇒ 00:29:50.800 Gabriel Lam: Consistent sometimes, compared to maybe some other ones.

317 00:29:51.290 ⇒ 00:29:51.909 Mustafa Raja: Oh, is…

318 00:29:51.910 ⇒ 00:29:53.070 Gabriel Lam: Where they’re a little more…

319 00:29:54.380 ⇒ 00:30:06.890 Mustafa Raja: Yeah, so this happens when… so we only updated the prompt for ticket generator and not the groomer. Okay. So, if a ticket actually belongs to a client, it’s going to go through another

320 00:30:06.940 ⇒ 00:30:16.580 Mustafa Raja: Prompt that we need to also update, and then it’s going to be consistent across all, descriptions, this format.

321 00:30:17.220 ⇒ 00:30:17.800 Gabriel Lam: Okay.

322 00:30:17.920 ⇒ 00:30:20.579 Gabriel Lam: Yeah, something I noticed, Mustafa, was

323 00:30:20.890 ⇒ 00:30:24.439 Gabriel Lam: When we… if we push this here, it’ll send it to the backlog.

324 00:30:24.560 ⇒ 00:30:30.700 Gabriel Lam: as the… the status. So maybe something there to include as well.

325 00:30:31.280 ⇒ 00:30:34.269 Gabriel Lam: Just so we… I think we already have a backlog. Yeah, what’s that.

326 00:30:34.270 ⇒ 00:30:41.429 Mustafa Raja: Yes, I will work on the link, link, attachment thing, and also, the status thing also.

327 00:30:41.900 ⇒ 00:30:44.920 Gabriel Lam: Okay, awesome. So something that…

328 00:30:45.230 ⇒ 00:30:49.060 Gabriel Lam: We were talking about was having… these…

329 00:30:49.400 ⇒ 00:30:56.589 Gabriel Lam: summaries be reflected into the stand-up assistant, just so instead of having to go through every meeting, you’ll be able to

330 00:30:57.030 ⇒ 00:31:01.770 Gabriel Lam: See the updates as a sort of live document here as a source of truth.

331 00:31:02.110 ⇒ 00:31:08.250 Gabriel Lam: we were hoping that these linear tickets would have some sort of human review, just because…

332 00:31:09.080 ⇒ 00:31:13.669 Gabriel Lam: The ability to update existing tickets is going to be a pretty heavy lift.

333 00:31:14.400 ⇒ 00:31:25.390 Gabriel Lam: So at least maybe having a Slack notification or some way to have a PM or person go through these tickets, just to be like, this is fine, this is incorrect, this is something we don’t need.

334 00:31:25.770 ⇒ 00:31:29.929 Gabriel Lam: And now that these persist, we’ll be able to see them,

335 00:31:31.640 ⇒ 00:31:34.850 Gabriel Lam: So, that’s sort of where we are. There are a couple things that…

336 00:31:35.110 ⇒ 00:31:37.360 Gabriel Lam: We are hoping to address today.

337 00:31:37.870 ⇒ 00:31:43.630 Gabriel Lam: Mainly to do with… the… the flow?

338 00:31:43.830 ⇒ 00:31:50.999 Gabriel Lam: And making sure that these tickets are usable and readable still. I think there’s some more testing that can be done, to match

339 00:31:51.230 ⇒ 00:31:55.610 Gabriel Lam: like, a sort of standard for everyone. I do notice everyone uses linear differently.

340 00:31:55.710 ⇒ 00:32:04.029 Gabriel Lam: But hoping that Exercise will have some standardization in, like, how people will read or receive or link tickets.

341 00:32:04.360 ⇒ 00:32:14.379 Gabriel Lam: We know that linear also allows you to, you know, link pull requests, link Slack messages, and so we’re also hoping to add, linkages there.

342 00:32:16.040 ⇒ 00:32:19.509 Gabriel Lam: So that’s sort of the general rundown.

343 00:32:19.510 ⇒ 00:32:20.510 Uttam Kumaran: Nice, guys.

344 00:32:20.510 ⇒ 00:32:24.870 Gabriel Lam: next week… sorry, I’ll just end with this. I think for next week, we also talked about…

345 00:32:25.130 ⇒ 00:32:37.090 Gabriel Lam: updating the client hub, as well as, moving the entire workstream into Mastra. I think maybe with next week being a lighter week, it could be something to think about.

346 00:32:37.520 ⇒ 00:32:38.900 Gabriel Lam: Just a thought.

347 00:32:41.830 ⇒ 00:32:47.810 Uttam Kumaran: Yeah, I’m game for that. I would prefer… yeah, actually, that’s probably a good…

348 00:32:48.370 ⇒ 00:32:59.520 Uttam Kumaran: game plan, because client work is gonna be a little bit light, and I know Casey Mustafa, unless you guys are celebrating Thanksgiving, but if you wanna keep… if you wanna work on stuff.

349 00:32:59.840 ⇒ 00:33:06.660 Uttam Kumaran: Yeah, that would be great. Yeah, maybe,

350 00:33:07.860 ⇒ 00:33:12.050 Uttam Kumaran: Maybe me and you can spend time today, Gabe,

351 00:33:12.500 ⇒ 00:33:19.680 Uttam Kumaran: I want to kind of spend the next… 30 minutes on… on blocking,

352 00:33:20.000 ⇒ 00:33:27.660 Uttam Kumaran: Sam and Mustafa on a couple other things, but maybe we can spend 30 minutes today?

353 00:33:29.020 ⇒ 00:33:32.050 Uttam Kumaran: If that’s fine… I… Yeah, that’s good.

354 00:33:32.710 ⇒ 00:33:39.630 Uttam Kumaran: I have this, like, hour-long partnership meeting that I may…

355 00:33:40.510 ⇒ 00:33:43.210 Uttam Kumaran: I may join and then leave.

356 00:33:45.890 ⇒ 00:33:46.770 Uttam Kumaran: Because…

357 00:33:46.770 ⇒ 00:33:50.380 Gabriel Lam: So you have, like… 3PM Eastern, or…

358 00:33:50.530 ⇒ 00:33:53.160 Gabriel Lam: I don’t know if you’re having lunch then, so I don’t wanna…

359 00:33:54.190 ⇒ 00:33:59.220 Uttam Kumaran: Yeah, I’m gonna try to get out of the house for, like, those end-of-the-day meetings.

360 00:33:59.720 ⇒ 00:34:00.310 Gabriel Lam: Yep.

361 00:34:00.680 ⇒ 00:34:09.400 Uttam Kumaran: It’s also, like, my… yeah, you’re just gonna catch me at the end of the week, at the end of, like, one of the, like, worst weeks ever, so it’s not…

362 00:34:09.840 ⇒ 00:34:10.750 Uttam Kumaran: I’m gonna be…

363 00:34:10.750 ⇒ 00:34:11.170 Samuel Roberts: reasonable.

364 00:34:11.179 ⇒ 00:34:20.359 Uttam Kumaran: So let’s plan on, like, can we plan on 10.30 Central, so 11.30? And I’ll… I’ll… I’m just gonna leave that meeting I’m in.

365 00:34:20.949 ⇒ 00:34:24.839 Uttam Kumaran: That would be great.

366 00:34:29.339 ⇒ 00:34:39.999 Uttam Kumaran: And then, yeah, if we need to do another… if we do need to do it later, too, we could do that, so… Okay. Perfect. And then, how do you feel, Casey, for the text-to-SQL

367 00:34:41.839 ⇒ 00:34:51.229 Uttam Kumaran: spike later. You don’t, of course, don’t have to be, like, sort of at any, like, finishing point, but just confirming you feel good about conducting that.

368 00:34:52.639 ⇒ 00:34:59.549 Casie Aviles: Yeah, I think I can have something for later, I’m just finalizing, like, Next steps…

369 00:35:00.040 ⇒ 00:35:00.370 Uttam Kumaran: Okay.

370 00:35:00.370 ⇒ 00:35:05.070 Casie Aviles: for, like… you know, the recommendations, I guess, for, like, the spike.

371 00:35:05.640 ⇒ 00:35:11.040 Casie Aviles: Yesterday, I did meet with… Clint from buying…

372 00:35:11.650 ⇒ 00:35:14.829 Casie Aviles: And I also added what I’ve…

373 00:35:15.130 ⇒ 00:35:17.940 Casie Aviles: my findings into that doc, so I think…

374 00:35:18.390 ⇒ 00:35:21.640 Casie Aviles: Should be able to get something reviewed later, and…

375 00:35:23.130 ⇒ 00:35:26.840 Casie Aviles: iterate if, you know, for any feedback that I get.

376 00:35:27.510 ⇒ 00:35:33.649 Uttam Kumaran: Okay. I guess my… my one is, like, if you can send The dock out before?

377 00:35:34.020 ⇒ 00:35:38.670 Uttam Kumaran: people can take a… take a look, especially because I won’t… I will only be on the first half hour.

378 00:35:38.800 ⇒ 00:35:43.720 Uttam Kumaran: So I just want to make sure to leave comments, and then, like, I can get any questions out.

379 00:35:43.830 ⇒ 00:35:45.789 Uttam Kumaran: And then…

380 00:35:46.780 ⇒ 00:35:57.070 Uttam Kumaran: Yeah, I would… I would just open it up if anyone else wants to join, like, I know maybe Amber could… may be interested, or Zoran, but the crew there is, like, the core crew, so…

381 00:35:57.810 ⇒ 00:36:01.519 Uttam Kumaran: Yeah, okay, that’s all I had for that.

382 00:36:02.080 ⇒ 00:36:02.770 Casie Aviles: Okay.

383 00:36:05.510 ⇒ 00:36:09.550 Uttam Kumaran: Okay, cool. So maybe we can use the,

384 00:36:10.970 ⇒ 00:36:18.959 Uttam Kumaran: rest of this, like, 30 minutes to work on, the Snowflake and dbt, sort of.

385 00:36:19.480 ⇒ 00:36:27.460 Uttam Kumaran: SOPs for initialization. I think, you know, anyone’s free to stay on, but it’ll mainly be me, Sam.

386 00:36:27.720 ⇒ 00:36:36.360 Uttam Kumaran: Mustafa, I guess, Awash, if you want to stay on, too, or if you’re just working, you want to listen in, basically, I’m… I want to write, like,

387 00:36:36.630 ⇒ 00:36:40.419 Uttam Kumaran: SOP for how to set up Snowflake, how to set up dbt.

388 00:36:40.740 ⇒ 00:36:44.189 Uttam Kumaran: I think me and you are mainly just doing it from…

389 00:36:44.470 ⇒ 00:36:47.180 Uttam Kumaran: I mean, from my muscle memory, so…

390 00:36:47.670 ⇒ 00:36:49.050 Awaish Kumar: Okay.

391 00:36:50.140 ⇒ 00:36:50.690 Uttam Kumaran: Cool.

392 00:36:51.450 ⇒ 00:36:52.790 Uttam Kumaran: Okay, alright.

393 00:36:54.570 ⇒ 00:36:56.970 Uttam Kumaran: Let me just… Cheer…

394 00:37:13.020 ⇒ 00:37:13.780 Uttam Kumaran: Okay.

395 00:37:14.830 ⇒ 00:37:20.270 Uttam Kumaran: So, to start this…

396 00:37:21.840 ⇒ 00:37:27.030 Uttam Kumaran: Let me just kind of think about starting this, and then I may just go make a quick coffee.

397 00:37:27.810 ⇒ 00:37:33.770 Uttam Kumaran: But let me pull up… Notion, so…

398 00:37:34.370 ⇒ 00:37:41.470 Uttam Kumaran: There’s kind of, like, two things I want to think about here as we’re starting to get to…

399 00:37:41.630 ⇒ 00:37:45.700 Uttam Kumaran: More, like, enterprise-level clients.

400 00:37:46.140 ⇒ 00:37:54.800 Uttam Kumaran: One thing that Robert sent me… was… Like…

401 00:37:55.700 ⇒ 00:37:59.200 Uttam Kumaran: That, you can’t see this, okay, hold on one sec.

402 00:38:02.700 ⇒ 00:38:05.359 Uttam Kumaran: Was, like, this stock.

403 00:38:06.820 ⇒ 00:38:12.340 Uttam Kumaran: And he was like, it’s kind of probably time for us to start thinking about, like, enterprise-level

404 00:38:12.460 ⇒ 00:38:14.150 Uttam Kumaran: materials like this.

405 00:38:14.930 ⇒ 00:38:17.289 Uttam Kumaran: This is just, like, a sample.

406 00:38:17.960 ⇒ 00:38:25.899 Uttam Kumaran: But, like, this is how for… Google’s… Marketing platform implementation.

407 00:38:27.500 ⇒ 00:38:31.389 Uttam Kumaran: This is sort of, like, how they create, like, a concise document around the whole thing.

408 00:38:33.290 ⇒ 00:38:38.090 Uttam Kumaran: I’m not as interested in this until we have SOPs on each thing.

409 00:38:38.400 ⇒ 00:38:43.530 Uttam Kumaran: But I do want to share that, like, this is the direction that we’re going.

410 00:38:44.510 ⇒ 00:38:53.190 Uttam Kumaran: And in particular, Element and these types of large clients expect this type of stuff from us.

411 00:38:53.620 ⇒ 00:38:58.230 Uttam Kumaran: So, maybe to start, and I can kind of show you guys, like.

412 00:38:58.450 ⇒ 00:39:00.950 Uttam Kumaran: maybe I’ll show you my process of, like.

413 00:39:01.310 ⇒ 00:39:08.970 Uttam Kumaran: how I would tackle this… this problem, and that way it gives you… gives you guys a little bit of insight. I’m not actually… of course.

414 00:39:09.550 ⇒ 00:39:13.390 Uttam Kumaran: surprise, I’m not gonna write the whole thing, so…

415 00:39:13.850 ⇒ 00:39:24.070 Uttam Kumaran: I… I will put the scaffold… let’s… let’s… let’s, as a team, work on, like, the scaffolding, and then I may, ask y’all to help.

416 00:39:24.360 ⇒ 00:39:25.270 Uttam Kumaran: with…

417 00:39:26.200 ⇒ 00:39:35.100 Uttam Kumaran: filling out some details. But let me show you, like, how I would start this. So, the first thing I… I mean, I’m gonna do is, like, I would just go to ChatGPT and…

418 00:39:35.230 ⇒ 00:39:41.425 Uttam Kumaran: I’m gonna go to… Our…

419 00:39:42.880 ⇒ 00:39:46.259 Uttam Kumaran: There’s, like, a CTO prompt somewhere.

420 00:39:49.100 ⇒ 00:39:51.730 Uttam Kumaran: Oh, CTO and ENG Operations Leader.

421 00:39:51.930 ⇒ 00:39:54.209 Uttam Kumaran: And I don’t remember, what was this prompt?

422 00:39:57.300 ⇒ 00:39:58.510 Uttam Kumaran: Okay, yeah.

423 00:39:58.810 ⇒ 00:40:01.450 Uttam Kumaran: So, basically.

424 00:40:02.200 ⇒ 00:40:09.729 Uttam Kumaran: I mean, I’m just gonna kinda relay, like, what it is we want to do to this guy, but I… I’m also gonna use Whisper, so I’ll just talk…

425 00:40:09.840 ⇒ 00:40:13.810 Uttam Kumaran: To it, so… We want to make,

426 00:40:14.720 ⇒ 00:40:27.699 Uttam Kumaran: a few documents today, but in particular, we have kind of several clients now that want to see an end-to-end SOP for Snowflake initialization, as well as DBT initialization.

427 00:40:28.150 ⇒ 00:40:38.739 Uttam Kumaran: Additionally, we should probably, in this process, write, an SOP for polyatomic initialization, Fivetran initialization.

428 00:40:41.740 ⇒ 00:40:47.670 Uttam Kumaran: Mother dock initialization, but dbt and Snowflake are the primary goals today.

429 00:40:47.840 ⇒ 00:40:49.490 Uttam Kumaran: Basically, I need help…

430 00:40:50.130 ⇒ 00:40:57.629 Uttam Kumaran: thinking about what the format for that SOP or runbook is gonna be, and this is gonna go into Notion to start.

431 00:40:58.390 ⇒ 00:41:02.819 Uttam Kumaran: And then eventually… may end up in a Google Sheet.

432 00:41:03.150 ⇒ 00:41:06.170 Uttam Kumaran: But Notion to start is fine.

433 00:41:06.490 ⇒ 00:41:09.180 Uttam Kumaran: So, ideally, this should be something that…

434 00:41:09.350 ⇒ 00:41:14.860 Uttam Kumaran: A mid-level engine… non-data engineer should be able to execute, so that’s kind of the level.

435 00:41:15.100 ⇒ 00:41:23.809 Uttam Kumaran: We already have documents on how to structure dbt, we have some boilerplate repos on…

436 00:41:24.140 ⇒ 00:41:28.880 Uttam Kumaran: BBT. We also have some scripts on grants for Snowflake.

437 00:41:29.070 ⇒ 00:41:35.090 Uttam Kumaran: But let’s just start with, like, that objective, and then you tell me What you need from me.

438 00:41:35.880 ⇒ 00:41:37.539 Uttam Kumaran: So this is how I…

439 00:41:38.560 ⇒ 00:41:42.639 Uttam Kumaran: AI, like, every day. So I come in, and I just hit whisper, and I just talk.

440 00:41:43.280 ⇒ 00:41:48.699 Uttam Kumaran: This would have been very hard to write, so as I said, told everyone here that I’m a big fan of Whisper, because…

441 00:41:49.190 ⇒ 00:41:51.500 Uttam Kumaran: Like, I could just do this really quickly.

442 00:41:51.820 ⇒ 00:41:57.140 Uttam Kumaran: And so, let’s just… Go ahead.

443 00:42:00.040 ⇒ 00:42:13.920 Uttam Kumaran: do what it says. In this, moment, I’m also going to be… putting together… This documentation…

444 00:42:14.250 ⇒ 00:42:25.980 Uttam Kumaran: under… How to… going to be… How to initialize… Snowflake…

445 00:42:28.130 ⇒ 00:42:32.099 Uttam Kumaran: Totally went somewhere else. How to initialize…

446 00:42:36.490 ⇒ 00:42:37.600 Uttam Kumaran: Snowflake.

447 00:42:39.880 ⇒ 00:42:41.770 Uttam Kumaran: Did that even get created?

448 00:42:49.360 ⇒ 00:42:53.969 Uttam Kumaran: Okay, I’m just gonna use this one, whatever.

449 00:42:55.580 ⇒ 00:43:01.710 Uttam Kumaran: how to… Initial, stuff like…

450 00:43:03.320 ⇒ 00:43:10.250 Uttam Kumaran: Okay, great, so… let’s see what it gave me… Here’s a concrete starting point.

451 00:43:11.090 ⇒ 00:43:22.270 Uttam Kumaran: And then how to initial… instantiate it for these. Generic Slack stack component initialization, overview, audience, Prereqs, inputs.

452 00:43:22.400 ⇒ 00:43:24.480 Uttam Kumaran: Outputs, time to complete.

453 00:43:25.050 ⇒ 00:43:29.700 Uttam Kumaran: Decision tree, search section at the top that tells the executor, is this a new account?

454 00:43:32.080 ⇒ 00:43:36.230 Uttam Kumaran: Alright, this seems pretty decent, it’s just gonna be a lot of work.

455 00:43:45.640 ⇒ 00:43:49.450 Uttam Kumaran: So, what do you guys think about… Like, this format.

456 00:43:51.910 ⇒ 00:43:57.550 Awaish Kumar: Yeah, I think there’s a lot of… Textual information, which…

457 00:43:57.950 ⇒ 00:44:01.899 Awaish Kumar: can be shortened? Like, we just… we should just have the…

458 00:44:02.450 ⇒ 00:44:07.280 Awaish Kumar: Basic introduction, and then the execution steps, like, to this.

459 00:44:07.280 ⇒ 00:44:07.660 Uttam Kumaran: Yeah.

460 00:44:10.190 ⇒ 00:44:14.860 Uttam Kumaran: Okay, so then I’m just gonna basically…

461 00:44:21.250 ⇒ 00:44:25.919 Uttam Kumaran: Yeah, I’m gonna put this… Like, this seems reasonable.

462 00:44:27.740 ⇒ 00:44:33.110 Uttam Kumaran: Inputs, prereqs, outputs… And then…

463 00:44:37.550 ⇒ 00:44:43.119 Uttam Kumaran: I guess I don’t really… We don’t need to have this decision tree, necessarily.

464 00:44:45.440 ⇒ 00:44:50.180 Uttam Kumaran: For each step summary, inside the toggle, steps validation, link screenshots.

465 00:45:01.260 ⇒ 00:45:06.570 Uttam Kumaran: Okay, so let’s, like, take a look at what it gave us for Snowflake. So, set up a secure…

466 00:45:06.880 ⇒ 00:45:09.889 Uttam Kumaran: output… Okay, great.

467 00:45:11.350 ⇒ 00:45:13.180 Uttam Kumaran: These are good outputs.

468 00:45:35.350 ⇒ 00:45:41.289 Uttam Kumaran: Okay, so let me just… I’m just gonna go ahead and copy these in, and then let’s… we just… let’s just get started on something.

469 00:45:42.860 ⇒ 00:45:46.299 Uttam Kumaran: You know, we can always change the format later.

470 00:45:49.960 ⇒ 00:45:50.820 Uttam Kumaran: Cool.

471 00:45:51.120 ⇒ 00:45:53.800 Uttam Kumaran: So I do feel like these are kind of the…

472 00:45:55.220 ⇒ 00:46:00.380 Uttam Kumaran: the steps. The other thing I want to do here…

473 00:46:03.170 ⇒ 00:46:06.920 Uttam Kumaran: Okay, so this is fine, confirm region, so…

474 00:46:10.640 ⇒ 00:46:16.609 Uttam Kumaran: I guess my other question is, like, should we copy a version of this for every client?

475 00:46:17.120 ⇒ 00:46:21.700 Uttam Kumaran: So it would be, like, go ahead and create a copy of this in the client Notion.

476 00:46:22.090 ⇒ 00:46:27.219 Uttam Kumaran: Because then what we can do at the top here is also say, like, put, like, a list of the details.

477 00:46:27.980 ⇒ 00:46:28.790 Uttam Kumaran: Right.

478 00:46:29.150 ⇒ 00:46:36.459 Uttam Kumaran: So, basically, for example, confirm region and cloud provider, you can list that here.

479 00:46:36.780 ⇒ 00:46:37.550 Uttam Kumaran: You know?

480 00:46:38.800 ⇒ 00:46:39.739 Samuel Roberts: Oh, I see.

481 00:46:40.860 ⇒ 00:46:41.660 Awaish Kumar: Okay.

482 00:46:41.940 ⇒ 00:46:46.909 Awaish Kumar: But, like, if we have something, like, general purpose, there can be…

483 00:46:47.050 ⇒ 00:46:52.360 Awaish Kumar: published as, like, in the Brainforge blog.

484 00:46:55.030 ⇒ 00:46:59.720 Uttam Kumaran: Yeah, like, we can make this public, because it doesn’t have any client info.

485 00:46:59.980 ⇒ 00:47:07.700 Uttam Kumaran: I guess there will be nuances, and I want to leave the client with this… Doc, ideally, you know?

486 00:47:10.390 ⇒ 00:47:20.970 Samuel Roberts: Yeah, I would think we’d want to have one that’s, like, generic, but with, like, a list of what you need to fill in, and whether or not we duplicate the Notion page, or just, like, copy that block out.

487 00:47:21.100 ⇒ 00:47:22.930 Samuel Roberts: And then follow the directions.

488 00:47:23.420 ⇒ 00:47:32.650 Samuel Roberts: So I don’t think we’re gonna need a separate auto-initialized snowflake for every client, necessarily, we just need the, like, what is the client-specific information that goes into this?

489 00:47:33.340 ⇒ 00:47:33.900 Uttam Kumaran: Yeah.

490 00:47:33.900 ⇒ 00:47:36.230 Samuel Roberts: And then you can reference that, I think.

491 00:47:36.810 ⇒ 00:47:37.430 Uttam Kumaran: Okay.

492 00:47:43.640 ⇒ 00:47:47.990 Uttam Kumaran: Okay, so let’s talk about… Each of these steps.

493 00:47:55.890 ⇒ 00:47:57.140 Uttam Kumaran: Secure…

494 00:48:16.950 ⇒ 00:48:23.260 Uttam Kumaran: Okay, so the first thing, I don’t know, Awash, maybe me and you could think through it. So, I mean, the first thing I do is…

495 00:48:24.470 ⇒ 00:48:30.529 Uttam Kumaran: I have… I mean, I just… we have, like, a script, so… I mainly follow that.

496 00:48:32.760 ⇒ 00:48:46.980 Uttam Kumaran: Which is… Somewhere… Like, here. So I follow, like, this script, so I’m gonna… but it’s actually not…

497 00:48:47.220 ⇒ 00:48:49.159 Uttam Kumaran: 100% correct anymore.

498 00:48:50.030 ⇒ 00:48:56.990 Uttam Kumaran: But… Yeah, like, so the first… I mean, the first piece I do is, like, I,

499 00:48:58.440 ⇒ 00:49:00.800 Uttam Kumaran: Basically, run this entire script.

500 00:49:04.540 ⇒ 00:49:08.240 Uttam Kumaran: I’m just gonna put this here, which is, like, Ron… this…

501 00:49:09.090 ⇒ 00:49:26.520 Uttam Kumaran: The… kind of, like, what this does… So, 1… Creates warehouses… Great. Databases… creates roles… Great.

502 00:49:27.250 ⇒ 00:49:41.290 Uttam Kumaran: Warehouses, databases, roles… And… runs… Grant…

503 00:49:45.750 ⇒ 00:49:46.910 Uttam Kumaran: Great.

504 00:49:47.120 ⇒ 00:49:53.879 Uttam Kumaran: service, account, users, So this script does all of this, not perfectly.

505 00:49:54.170 ⇒ 00:49:59.280 Uttam Kumaran: And so… I think the… kind of, like, what I do want to have in this…

506 00:49:59.480 ⇒ 00:50:06.089 Uttam Kumaran: Here is, like, actually each… Kind of an indication of, like, what’s happening, like, what this script does.

507 00:50:06.400 ⇒ 00:50:11.500 Uttam Kumaran: So, yeah, we have, like, some database… And,

508 00:50:12.650 ⇒ 00:50:15.530 Uttam Kumaran: You don’t really need to create a schema, actually.

509 00:50:18.440 ⇒ 00:50:28.490 Uttam Kumaran: So, we have raw… enter… Immediate… And then… March…

510 00:50:31.490 ⇒ 00:50:35.990 Uttam Kumaran: And then this one, yeah, we did create some standard rules.

511 00:50:38.620 ⇒ 00:50:45.040 Uttam Kumaran: Warehouses, grants… Yeah, this is a to-do, configure session timeout.

512 00:50:54.740 ⇒ 00:51:10.799 Uttam Kumaran: I know we don’t need to do this… And then… Here… We need to do… table of properties…

513 00:51:14.900 ⇒ 00:51:16.540 Uttam Kumaran: Checklist, basically.

514 00:51:17.710 ⇒ 00:51:24.220 Uttam Kumaran: So, so I think… It would be helpful for…

515 00:51:24.700 ⇒ 00:51:29.370 Uttam Kumaran: Mustafa and Sam, for you guys to familiarize yourself with this script.

516 00:51:30.060 ⇒ 00:51:32.750 Samuel Roberts: Because this is what we’re going… this is actually…

517 00:51:33.100 ⇒ 00:51:40.860 Uttam Kumaran: Encompasses a lot of… This encompasses a lot of the steps, actually.

518 00:51:41.450 ⇒ 00:51:43.829 Uttam Kumaran: So, Snowflake is funny because, like.

519 00:51:44.650 ⇒ 00:51:50.600 Uttam Kumaran: You can probably run this, but you may hit a snag, and so you still need to know what every step is doing.

520 00:51:51.220 ⇒ 00:51:51.580 Samuel Roberts: Right.

521 00:51:51.580 ⇒ 00:51:57.520 Uttam Kumaran: And I think Awash, I may need your help to update this script.

522 00:52:00.940 ⇒ 00:52:06.209 Uttam Kumaran: Because previously, I was creating a read…

523 00:52:07.130 ⇒ 00:52:11.750 Uttam Kumaran: a read-write role for every environment. That seems accurate, right?

524 00:52:14.510 ⇒ 00:52:16.230 Uttam Kumaran: but I don’t have roles…

525 00:52:16.370 ⇒ 00:52:17.290 Awaish Kumar: like…

526 00:52:18.120 ⇒ 00:52:26.640 Uttam Kumaran: We have role developer, I think we need to also have… Another, probably, role?

527 00:52:27.020 ⇒ 00:52:28.340 Uttam Kumaran: like, role…

528 00:52:28.690 ⇒ 00:52:39.119 Uttam Kumaran: Like, we have role analy… maybe we’re a role analyst, role developer, we need to have, like, a couple… we need to have a couple more, and then basically you grant… you do, like, you grant roles to the role.

529 00:52:39.710 ⇒ 00:52:43.790 Uttam Kumaran: So you consolidate them, and then also, like, these service keys…

530 00:52:44.330 ⇒ 00:52:47.100 Uttam Kumaran: You need to… you need to actually initialize…

531 00:52:47.260 ⇒ 00:52:49.640 Uttam Kumaran: The public keys before you do this.

532 00:52:50.630 ⇒ 00:52:52.669 Uttam Kumaran: So that’s also something here.

533 00:52:53.250 ⇒ 00:53:04.879 Uttam Kumaran: Like, for the service accounts… You need to initialize the… The private and public keys.

534 00:53:07.130 ⇒ 00:53:09.150 Uttam Kumaran: beforehand.

535 00:53:09.910 ⇒ 00:53:13.540 Uttam Kumaran: Otherwise, if you run this, you’ll create them, but there won’t be any public keys.

536 00:53:14.080 ⇒ 00:53:15.160 Uttam Kumaran: And then the last.

537 00:53:15.160 ⇒ 00:53:15.540 Samuel Roberts: interview.

538 00:53:15.540 ⇒ 00:53:16.150 Uttam Kumaran: Yeah.

539 00:53:16.950 ⇒ 00:53:22.380 Samuel Roberts: So is that… this… this SQL file gets run on Snowflake Online, or is it…

540 00:53:22.380 ⇒ 00:53:22.720 Uttam Kumaran: Yeah.

541 00:53:22.720 ⇒ 00:53:24.239 Samuel Roberts: you do in the CLI? Okay.

542 00:53:24.700 ⇒ 00:53:28.380 Uttam Kumaran: You can actually run this in the CL… You can run this in the CLI.

543 00:53:28.380 ⇒ 00:53:37.160 Samuel Roberts: That’s what I’m wondering, because I’m wondering if we could automate the creation of those keys and everything as part of, like, a bash script that runs this. I just don’t know, I’m not too familiar with how Snowflake…

544 00:53:37.720 ⇒ 00:53:38.869 Uttam Kumaran: No, there’s no CLI.

545 00:53:38.870 ⇒ 00:53:40.310 Samuel Roberts: thing. Okay.

546 00:53:40.310 ⇒ 00:53:46.290 Uttam Kumaran: Yeah, so… Yeah, I mean, see, this is why I bring in the smarter people.

547 00:53:47.010 ⇒ 00:53:51.630 Uttam Kumaran: Things get a lot better here. So, this is… yeah, Snowflake does have a CLI.

548 00:53:52.120 ⇒ 00:53:54.009 Uttam Kumaran: I don’t know…

549 00:53:58.890 ⇒ 00:54:00.370 Samuel Roberts: Yeah, I mean…

550 00:54:01.820 ⇒ 00:54:06.510 Uttam Kumaran: Totally would be worth… yeah, I’m just gonna… I’m gonna just put that in here somewhere.

551 00:54:07.160 ⇒ 00:54:07.960 Samuel Roberts: Yeah, yeah.

552 00:54:07.960 ⇒ 00:54:11.120 Uttam Kumaran: Maybe, like, a higher level to-do, which is one…

553 00:54:11.660 ⇒ 00:54:15.370 Uttam Kumaran: And we run this using the CLI.

554 00:54:16.510 ⇒ 00:54:17.989 Uttam Kumaran: That would be great.

555 00:54:18.970 ⇒ 00:54:26.829 Uttam Kumaran: And then second is, like, basically also, for most clients, we are initializing a repo, so you can initialize a repo.

556 00:54:27.190 ⇒ 00:54:30.320 Uttam Kumaran: add the script, and then run that via the CLI.

557 00:54:35.060 ⇒ 00:54:39.909 Samuel Roberts: Because, yeah, if there’s other things that have to get plugged in there besides keys, we could have that all, like, as inputs or something.

558 00:54:40.610 ⇒ 00:54:41.010 Uttam Kumaran: Great.

559 00:54:41.010 ⇒ 00:54:43.289 Samuel Roberts: The keys could get generated and saved, and…

560 00:54:44.150 ⇒ 00:54:44.740 Uttam Kumaran: Great.

561 00:54:55.640 ⇒ 00:54:56.580 Uttam Kumaran: Cool.

562 00:54:57.480 ⇒ 00:55:01.160 Uttam Kumaran: So this is a good start. I think I would probably…

563 00:55:02.280 ⇒ 00:55:08.980 Uttam Kumaran: ask, maybe, Mustafa, you and Sam to take a read of the… of the script to date.

564 00:55:09.470 ⇒ 00:55:10.810 Uttam Kumaran: And then…

565 00:55:11.570 ⇒ 00:55:20.249 Uttam Kumaran: I mean, my subjug… I’m still not happy with this format, but I think it’s, like, getting closer. Basically, we want it to be, like, really dummy-proof.

566 00:55:20.430 ⇒ 00:55:26.860 Uttam Kumaran: So, I think there’s gonna be a section on, like, what the script is doing, and then we need to have a section on, like, how to execute it.

567 00:55:27.310 ⇒ 00:55:27.970 Mustafa Raja: Hmm.

568 00:55:29.050 ⇒ 00:55:29.920 Uttam Kumaran: You know?

569 00:55:30.520 ⇒ 00:55:31.280 Uttam Kumaran: Yeah.

570 00:55:31.620 ⇒ 00:55:35.710 Uttam Kumaran: Or around, like, what is this script doing?

571 00:55:36.450 ⇒ 00:55:43.260 Uttam Kumaran: And then we kind of need, like, a section on, like, Like, step by step.

572 00:55:44.120 ⇒ 00:55:44.640 Samuel Roberts: Right.

573 00:55:44.640 ⇒ 00:55:45.390 Uttam Kumaran: execution.

574 00:55:46.140 ⇒ 00:55:51.479 Samuel Roberts: Now, is this this Notion thing, is this something that could live in the repo as, like, a…

575 00:55:51.480 ⇒ 00:55:53.940 Uttam Kumaran: Yeah, yeah, you could also do that, yeah, yeah.

576 00:55:53.940 ⇒ 00:55:54.590 Samuel Roberts: Okay.

577 00:55:54.770 ⇒ 00:56:01.390 Samuel Roberts: I didn’t know if it needed to be something that was, like, outside there just for, like, our purposes, but if it’s… if it’s gonna be something that we’re gonna be duplicating and keeping, like.

578 00:56:02.390 ⇒ 00:56:05.260 Uttam Kumaran: Yeah, that’s a great point also, yeah, this should just live there.

579 00:56:06.890 ⇒ 00:56:12.169 Samuel Roberts: Right, so, like, some kind of template repo that we have would add this as well, which I think is what we have right now, it’s just…

580 00:56:12.170 ⇒ 00:56:14.109 Uttam Kumaran: We already do have a template repo, so you should totally.

581 00:56:14.110 ⇒ 00:56:24.539 Samuel Roberts: Exactly, so I think we want to add this as, like, a README that, like, is tied to that script, probably, or something. Or that SQL script, but potentially also, like, a bash script, maybe.

582 00:56:25.050 ⇒ 00:56:25.390 Uttam Kumaran: Great.

583 00:56:25.390 ⇒ 00:56:25.970 Samuel Roberts: Okay.

584 00:56:26.220 ⇒ 00:56:27.030 Samuel Roberts: Cool.

585 00:56:28.210 ⇒ 00:56:30.070 Uttam Kumaran: Do you guys want to take a stab?

586 00:56:30.280 ⇒ 00:56:31.860 Uttam Kumaran: Add this from here.

587 00:56:33.030 ⇒ 00:56:36.049 Uttam Kumaran: I would say you just… just take a look at each of these.

588 00:56:36.050 ⇒ 00:56:36.660 Samuel Roberts: Yeah.

589 00:56:37.150 ⇒ 00:56:48.080 Uttam Kumaran: sections. Basically, role-based access control is, like, you want… the ideal situation is you have writer and reader roles for environments, and then those get granted to higher-level roles.

590 00:56:48.630 ⇒ 00:56:53.929 Uttam Kumaran: But you never grant… you never grant access directly to a user.

591 00:56:54.090 ⇒ 00:56:57.150 Uttam Kumaran: You’re always granting it to a role, and a user has a role.

592 00:56:57.860 ⇒ 00:56:58.380 Mustafa Raja: Yep.

593 00:56:58.380 ⇒ 00:56:59.869 Uttam Kumaran: That’s, like, the sort of the principle.

594 00:57:01.960 ⇒ 00:57:03.000 Samuel Roberts: Okay, yeah.

595 00:57:04.710 ⇒ 00:57:05.869 Uttam Kumaran: Okay, so if you guys wanna…

596 00:57:05.870 ⇒ 00:57:06.490 Samuel Roberts: and…

597 00:57:06.490 ⇒ 00:57:14.350 Uttam Kumaran: Take a stab at this, and then… yeah, maybe we can also talk briefly about…

598 00:57:16.080 ⇒ 00:57:19.010 Uttam Kumaran: Like… like, how to,

599 00:57:23.340 ⇒ 00:57:27.900 Uttam Kumaran: So, in this doc, let me… I’ll just paste in whatever it gave me.

600 00:57:35.030 ⇒ 00:57:35.860 Samuel Roberts: Oh, yeah.

601 00:57:37.160 ⇒ 00:57:45.899 Uttam Kumaran: Yeah, this is so… it’s right about dbt Cloud versus dbt Core.

602 00:57:46.240 ⇒ 00:57:51.990 Uttam Kumaran: New repo, and then, yeah, it’s… these are all gonna be, for the most part, a single environment, so…

603 00:57:52.880 ⇒ 00:57:59.020 Uttam Kumaran: yeah, these are the two kind of thing, dbt project, and then we also basically will have…

604 00:58:01.470 ⇒ 00:58:03.959 Uttam Kumaran: So this is a new, existing…

605 00:58:04.880 ⇒ 00:58:10.520 Uttam Kumaran: and new dbt cloud… so, new dbt… we won’t… we’re never gonna have a new dbt cloud.

606 00:58:10.840 ⇒ 00:58:17.749 Uttam Kumaran: But you may have an existing dbt project, so I’m gonna put in here… There’s, like…

607 00:58:20.360 ⇒ 00:58:25.930 Uttam Kumaran: This is gonna be dbt Core, and then this is gonna be new dbt Cloud.

608 00:58:27.110 ⇒ 00:58:29.139 Uttam Kumaran: So there’s gonna be steps there.

609 00:58:29.490 ⇒ 00:58:30.420 Uttam Kumaran: Yeah.

610 00:58:30.740 ⇒ 00:58:36.880 Uttam Kumaran: Already for this, we do have, our template repo…

611 00:58:45.860 ⇒ 00:58:46.770 Uttam Kumaran: Yes.

612 00:58:47.620 ⇒ 00:58:53.140 Uttam Kumaran: So, this is, like, a new client template. So, we kind of, like, honestly want to beef this up.

613 00:58:53.440 ⇒ 00:58:56.529 Uttam Kumaran: For sure, for sure. Everybody has access to this?

614 00:59:09.620 ⇒ 00:59:10.929 Uttam Kumaran: Yeah, I think so.

615 00:59:13.080 ⇒ 00:59:15.060 Samuel Roberts: which teams I’m on or anything, so…

616 00:59:16.010 ⇒ 00:59:20.759 Samuel Roberts: But I also have access, I can probably give myself access if I need, or tweak that if anyone doesn’t have it.

617 00:59:24.030 ⇒ 00:59:24.740 Uttam Kumaran: Okay.

618 00:59:28.210 ⇒ 00:59:39.490 Uttam Kumaran: Okay, everybody should have that. So, yeah, I think, kind of, like, we want to… in this repo, we do have initialized, like, a dbt project here, but if you’re using dbt Cloud, you don’t necessarily need to do this.

619 00:59:40.380 ⇒ 00:59:43.989 Uttam Kumaran: So there is some steps on dbt Cloud, which is, like, sign up.

620 00:59:44.520 ⇒ 00:59:45.070 Samuel Roberts: Yep.

621 00:59:45.550 ⇒ 00:59:52.289 Uttam Kumaran: Using the client domain… Email address, if possible.

622 00:59:53.230 ⇒ 01:00:00.810 Uttam Kumaran: And then there’s some steps in dbt Cloud for us to do, like… There’s, like, connections, there’s…

623 01:00:01.250 ⇒ 01:00:09.359 Uttam Kumaran: The dev environment, there’s… setting up… the first…

624 01:00:09.790 ⇒ 01:00:22.010 Uttam Kumaran: Job, testing, the first job. There’s dev environments, staging end, production, and… Adding more users…

625 01:00:23.010 ⇒ 01:00:25.740 Uttam Kumaran: Yeah, there’s a couple of steps here.

626 01:00:26.730 ⇒ 01:00:35.450 Uttam Kumaran: And so this is all gonna be within, like, mostly UI, and then these are gonna be all…

627 01:00:37.480 ⇒ 01:00:38.970 Uttam Kumaran: within cursor…

628 01:00:39.780 ⇒ 01:00:41.060 Samuel Roberts: But again, like…

629 01:00:41.210 ⇒ 01:00:46.680 Uttam Kumaran: Yeah, I’m saying, I mean, I’m… if you’re like, hey, we can do half this shit, like, just through a script, then we should just do that.

630 01:00:46.680 ⇒ 01:00:47.430 Awaish Kumar: Yeah.

631 01:00:47.570 ⇒ 01:00:49.430 Awaish Kumar: Setting up connections…

632 01:00:50.660 ⇒ 01:00:52.149 Uttam Kumaran: Yeah, yeah.

633 01:00:53.270 ⇒ 01:00:54.210 Samuel Roberts: Bye.

634 01:00:54.470 ⇒ 01:00:55.310 Awaish Kumar: Oh, okay.

635 01:00:57.260 ⇒ 01:00:59.589 Uttam Kumaran: Yeah, I’ll wrap my head around it a little bit and see what I… Okay.

636 01:00:59.790 ⇒ 01:01:02.920 Samuel Roberts: Understandable, like, what’s automatable a little more and what’s not.

637 01:01:03.500 ⇒ 01:01:14.759 Uttam Kumaran: And then let me create one thing here, which is gonna be, like… engineering… New client setup.

638 01:01:14.970 ⇒ 01:01:16.110 Uttam Kumaran: So, peace.

639 01:01:16.540 ⇒ 01:01:32.570 Uttam Kumaran: And I’m going to… Link… the… how to dbt… And I’m also gonna link… the how to snowflake…

640 01:01:37.570 ⇒ 01:01:42.730 Uttam Kumaran: And then in this one also, I want to link… Somewhere…

641 01:01:45.940 ⇒ 01:01:49.810 Uttam Kumaran: We… DVC… Great.

642 01:01:54.080 ⇒ 01:01:59.139 Uttam Kumaran: Yeah, so I think this is new client setup SOPs. I am going to…

643 01:02:00.520 ⇒ 01:02:04.900 Uttam Kumaran: I’m going to also share with y’all this…

644 01:02:05.250 ⇒ 01:02:07.920 Uttam Kumaran: I’m gonna put this ChatGPT thread…

645 01:02:08.290 ⇒ 01:02:11.350 Uttam Kumaran: Into here, so anyone can leverage that.

646 01:02:13.690 ⇒ 01:02:14.200 Samuel Roberts: Cool.

647 01:02:23.630 ⇒ 01:02:26.090 Uttam Kumaran: So you can start from here if you need it.

648 01:02:28.440 ⇒ 01:02:29.390 Uttam Kumaran: Q.

649 01:02:35.330 ⇒ 01:02:41.140 Uttam Kumaran: And then… The to-dos here is, like, one, create and…

650 01:02:41.590 ⇒ 01:02:45.400 Uttam Kumaran: SOP library for new client setup.

651 01:02:46.770 ⇒ 01:02:58.019 Uttam Kumaran: And then… Sop library for a new client setup, potentially add to…

652 01:02:58.450 ⇒ 01:03:02.340 Uttam Kumaran: Data, platform, Google Sheets? Question mark?

653 01:03:02.680 ⇒ 01:03:04.580 Uttam Kumaran: Update template.

654 01:03:04.980 ⇒ 01:03:05.900 Uttam Kumaran: repo.

655 01:03:07.370 ⇒ 01:03:11.890 Uttam Kumaran: Okay. Alright, so yeah, Sam, I guess if you guys want to take a look at that, that’d be great.

656 01:03:12.370 ⇒ 01:03:15.929 Uttam Kumaran: And because we’ll be, yeah, running that for a few clients then.

657 01:03:17.470 ⇒ 01:03:20.689 Uttam Kumaran: Okay, cool.

658 01:03:20.980 ⇒ 01:03:26.100 Uttam Kumaran: I feel like, let’s see who joins this… The next call…

659 01:03:31.340 ⇒ 01:03:34.489 Uttam Kumaran: I’m going to quickly run and get a coffee.

660 01:03:34.660 ⇒ 01:03:36.799 Uttam Kumaran: Just for my kitchen. One second.

661 01:03:37.570 ⇒ 01:03:38.750 Uttam Kumaran: I’ll be back.

662 01:03:39.220 ⇒ 01:03:41.940 Uttam Kumaran: But maybe, Awash, if you wanna…

663 01:03:42.250 ⇒ 01:03:46.160 Uttam Kumaran: Pass anything to Ashwini for Eden, you guys can start with that.

664 01:03:48.700 ⇒ 01:03:51.049 Samuel Roberts: Do you want me on this next call, or do you want me to jump into this now?

665 01:03:51.710 ⇒ 01:03:52.829 Samuel Roberts: Am I already begun?

666 01:03:57.470 ⇒ 01:03:58.940 Awaish Kumar: Hello, Ashwini.

667 01:03:58.940 ⇒ 01:03:59.780 Ashwini Sharma: Anyway…

668 01:04:01.640 ⇒ 01:04:03.080 Awaish Kumar: I, yeah, like…

669 01:04:03.300 ⇒ 01:04:09.369 Awaish Kumar: For now, I think you have a… like, we merged the PR yesterday, I think it went…

670 01:04:09.720 ⇒ 01:04:17.830 Awaish Kumar: Well, so, I think for today, like, we have a ticket for Metaplane.

671 01:04:18.300 ⇒ 01:04:22.419 Ashwini Sharma: Yes, I’m… I’m working on that, ticket.

672 01:04:23.400 ⇒ 01:04:24.370 Ashwini Sharma: Yeah, I mean…

673 01:04:25.100 ⇒ 01:04:32.440 Awaish Kumar: So you can look at the tables in BigQuery, and you can also familiarize yourself with Metaplane.

674 01:04:32.640 ⇒ 01:04:35.999 Awaish Kumar: Tool, and there are existing monitors.

675 01:04:36.290 ⇒ 01:04:40.100 Awaish Kumar: And how they are set up, you can just familiarize yourself with that.

676 01:04:40.210 ⇒ 01:04:42.110 Awaish Kumar: And then you can,

677 01:04:42.940 ⇒ 01:04:52.850 Awaish Kumar: like, like, we need, like, number one thing is we need coverage, like, we added quite a few tables, afterwards, and,

678 01:04:53.180 ⇒ 01:05:01.799 Awaish Kumar: But we didn’t really have monitors for all of them. So, number one thing is that we need to cover all of those. Secondly,

679 01:05:01.940 ⇒ 01:05:08.190 Awaish Kumar: We… the only thing to consider is that we are only adding monitors for March table.

680 01:05:08.480 ⇒ 01:05:17.790 Awaish Kumar: So, only those tables which are in bars, so we don’t… right now, we are not concerned about the tables in intermediate or raw tables.

681 01:05:19.420 ⇒ 01:05:23.820 Ashwini Sharma: Right, yeah, so, yeah, I just had a question regarding that, not…

682 01:05:24.150 ⇒ 01:05:27.800 Ashwini Sharma: I mean, like, it’s just a thought process in my mind, right? Maybe…

683 01:05:27.950 ⇒ 01:05:39.449 Ashwini Sharma: somebody who is already working with Metaplane could answer that. You know, I found that work sort of repetitive, right? So, for example, like, if we have 50 tables that are really important, and we need to monitor them.

684 01:05:39.920 ⇒ 01:05:43.670 Ashwini Sharma: Let’s say it’s only for freshness, right? And…

685 01:05:43.810 ⇒ 01:05:51.069 Ashwini Sharma: what it means is, like, you have to do this task 50 times for 50 tables, right? When you’re doing it from the UI, or…

686 01:05:51.190 ⇒ 01:05:53.360 Ashwini Sharma: I don’t know if there is a better way to do it.

687 01:05:54.430 ⇒ 01:06:10.310 Awaish Kumar: Like, you can add, like… so there are different ways in Metaplane itself that you can basically add all the monitors on all the tables which are frequently accessed, or you can even set up the rule.

688 01:06:10.800 ⇒ 01:06:14.939 Awaish Kumar: Which says, any, like, you can maybe set up, like.

689 01:06:15.120 ⇒ 01:06:22.909 Awaish Kumar: Any table in this schema should automatically have Freshness Monitor. You can set those rules in MetaPlayer.

690 01:06:23.680 ⇒ 01:06:24.450 Ashwini Sharma: Okay.

691 01:06:27.340 ⇒ 01:06:36.989 Awaish Kumar: So the next time, whenever there is a table created in, for example, broad marks, we’ll have that monitor. So right now, we just added

692 01:06:37.340 ⇒ 01:06:44.820 Awaish Kumar: And even then, like, we had quite a lot of March tables, so we slimmed down to quite a

693 01:06:45.030 ⇒ 01:06:51.719 Awaish Kumar: To a few tables, initially, so that we are not get… we don’t get spammed by…

694 01:06:51.950 ⇒ 01:06:56.639 Awaish Kumar: We don’t have alerts, but right now, like, we have… we are good enough people here.

695 01:06:56.760 ⇒ 01:07:03.760 Awaish Kumar: to cater that, we can go with those rules, like, we want to cover, all the tables in ProdMars.

696 01:07:05.900 ⇒ 01:07:11.930 Ashwini Sharma: Okay, yeah, yeah, sure, sure. Yeah, I have that list based on how frequently each table is accessed.

697 01:07:12.140 ⇒ 01:07:17.829 Ashwini Sharma: Based on that, I’m creating those monitors, just doing it one by one.

698 01:07:18.530 ⇒ 01:07:25.469 Ashwini Sharma: Let me see if there is a better way to do it, like, if we can directly create all the freshness monitor for all the tables.

699 01:07:25.470 ⇒ 01:07:26.260 Awaish Kumar: Yup.

700 01:07:27.840 ⇒ 01:07:28.990 Awaish Kumar: I think that’s…

701 01:07:29.420 ⇒ 01:07:36.109 Awaish Kumar: For freshness, then we just want to add some for revenue. Like, there are some monitors, and there are…

702 01:07:36.440 ⇒ 01:07:42.449 Awaish Kumar: Grow monitors, so we just want to add a few of them.

703 01:07:42.660 ⇒ 01:07:46.849 Awaish Kumar: On the, on the really critical models, like.

704 01:07:47.100 ⇒ 01:07:53.610 Ashwini Sharma: So, what are the most important monitoring metrics? Like, one is freshness.

705 01:07:53.610 ⇒ 01:08:02.899 Awaish Kumar: Yes, right. The second thing is that, nothing’s refreshed, but the,

706 01:08:03.400 ⇒ 01:08:15.729 Awaish Kumar: But, like, the revenues, like, the rows are same, or sometimes they are, like, there are duplicates, and we get extra rows, or some of revenue gets, like,

707 01:08:16.399 ⇒ 01:08:22.160 Awaish Kumar: Makeup’s really high, so, like… Things like that. So we need to add some…

708 01:08:22.439 ⇒ 01:08:28.770 Awaish Kumar: some revenues, some, like, raw monitors, some freshness monitors. I think that’s all for now.

709 01:08:29.870 ⇒ 01:08:30.510 Ashwini Sharma: Okay.

710 01:08:30.510 ⇒ 01:08:37.109 Awaish Kumar: And there are some… then we can set up some SLAs, right, in our monitors, freshness monitors.

711 01:08:37.270 ⇒ 01:08:41.869 Awaish Kumar: We don’t want to, like, use… there are…

712 01:08:41.939 ⇒ 01:08:57.900 Awaish Kumar: different configurations you can set up, like automatic or manual. Automatic auto, like, training of the model, and figures to figure out what is the normal SLA for this table to be… to get fresh… refreshed.

713 01:08:57.899 ⇒ 01:09:03.800 Awaish Kumar: Instead, we can set up some manual SLAs, like, we know, for example, that…

714 01:09:03.890 ⇒ 01:09:06.319 Awaish Kumar: For Eden, we know that,

715 01:09:07.140 ⇒ 01:09:12.519 Awaish Kumar: So they, they see the data, like, Every day.

716 01:09:12.720 ⇒ 01:09:23.200 Awaish Kumar: But, like, so if a table… and our… basically, our models run every hour, so if we say.

717 01:09:23.609 ⇒ 01:09:30.299 Awaish Kumar: That, like, A table gets refreshed… refreshed in, like, a…

718 01:09:31.000 ⇒ 01:09:36.860 Awaish Kumar: In a 6-hour span is good enough, Like, it…

719 01:09:37.580 ⇒ 01:09:42.150 Awaish Kumar: it’s good enough, so we don’t have to send alerts for those. So, for example.

720 01:09:42.200 ⇒ 01:09:59.040 Awaish Kumar: In a day, we are refreshing model every hour, but it is possible that in that hour, we didn’t get any new spend, new added spend, and the number of row counts remains the same.

721 01:09:59.880 ⇒ 01:10:03.089 Awaish Kumar: But that means… but that is still valid?

722 01:10:04.310 ⇒ 01:10:07.729 Awaish Kumar: Because we might not spend in the last few hours.

723 01:10:08.510 ⇒ 01:10:13.629 Awaish Kumar: Things like that. So you just have to consider these SLAs.

724 01:10:14.590 ⇒ 01:10:22.019 Ashwini Sharma: So, who decides when, when, when, I mean, how frequently the model gets refreshed? Is it the customer? Eden?

725 01:10:22.920 ⇒ 01:10:26.930 Awaish Kumar: Yeah, we basically run it every hour.

726 01:10:27.700 ⇒ 01:10:30.090 Awaish Kumar: Right now, because Eden wanted…

727 01:10:30.740 ⇒ 01:10:46.650 Awaish Kumar: like, some of the team members in the Eden, they… like, they’re the analysts, they were doing some manual carries, and they needed some, data to be refreshed. So we ended up, like, refreshing it every hour.

728 01:10:47.620 ⇒ 01:10:49.170 Awaish Kumar: Yeah.

729 01:10:52.680 ⇒ 01:10:57.100 Ashwini Sharma: And how often does the source get referenced? The raw layer, or…

730 01:10:57.880 ⇒ 01:11:02.229 Awaish Kumar: Yeah, like, that is… that is also, like, the same, because…

731 01:11:02.510 ⇒ 01:11:05.319 Awaish Kumar: Raw layer is, like, even distributed.

732 01:11:06.090 ⇒ 01:11:17.610 Awaish Kumar: So, in the event is streaming, so Basque basically gets, like, the Eden gets some orders through their platform called Bask, and then they are streamed into our

733 01:11:18.010 ⇒ 01:11:21.460 Awaish Kumar: BigQuery, basically.

734 01:11:21.720 ⇒ 01:11:27.109 Awaish Kumar: But that’s, like, maybe, you know… that also is, like, takes one hour.

735 01:11:29.440 ⇒ 01:11:30.210 Ashwini Sharma: Okay.

736 01:11:32.700 ⇒ 01:11:34.820 Awaish Kumar: to get new events in, yeah.

737 01:11:39.150 ⇒ 01:11:45.350 Ashwini Sharma: And this is happening directly through GitHub Actions, right? That’s the only orchestration?

738 01:11:46.330 ⇒ 01:11:53.480 Awaish Kumar: So, ingestion is… ingestion is happening through multiple tools, so I have a…

739 01:11:53.870 ⇒ 01:12:01.369 Awaish Kumar: Google Sheet I can share with you. It will… just gives you the summary of the tools we are using, but…

740 01:12:01.990 ⇒ 01:12:06.589 Awaish Kumar: ingestion happens using few tools like polyatomic, Segment.

741 01:12:06.910 ⇒ 01:12:14.619 Awaish Kumar: Google Cloud Function, Dexter, but, then, the modeling…

742 01:12:15.440 ⇒ 01:12:20.699 Awaish Kumar: It is done through dbt, and all the dbt modules are actually running through… only through Git of Action.

743 01:12:27.510 ⇒ 01:12:28.370 Ashwini Sharma: Okay.

744 01:12:29.400 ⇒ 01:12:32.639 Awaish Kumar: So, transformation layer is GitHub Actions.

745 01:12:33.280 ⇒ 01:12:37.710 Awaish Kumar: Ingestion is spread across multiple tools because of various reasons.

746 01:12:37.840 ⇒ 01:12:39.330 Awaish Kumar: And then, yeah.

747 01:12:39.830 ⇒ 01:12:42.670 Awaish Kumar: Visualization happens in Tableau.

748 01:12:43.890 ⇒ 01:12:44.780 Awaish Kumar: Excellent.

749 01:12:48.440 ⇒ 01:12:53.009 Ashwini Sharma: And is this the standard that we follow across all the clients, or…

750 01:12:53.730 ⇒ 01:12:56.579 Awaish Kumar: No, it depends on the client, right? So…

751 01:12:57.400 ⇒ 01:13:10.460 Awaish Kumar: For this client, we basically inherited some of the tools they were using, like Segment. They are pretty much on Segment. They use it for different purposes than just using it for analytics.

752 01:13:10.590 ⇒ 01:13:19.690 Awaish Kumar: Like, they’re… They use it to… for the identity stretching, and… and trying to be the…

753 01:13:19.850 ⇒ 01:13:24.520 Awaish Kumar: HIPAA compliant and things like that, so they use Segment for that.

754 01:13:25.500 ⇒ 01:13:29.429 Awaish Kumar: And they used… they already used… were using BigQuery.

755 01:13:29.680 ⇒ 01:13:35.189 Awaish Kumar: So, whether we introduce some tools, like Polytomic, Daxter.

756 01:13:43.020 ⇒ 01:13:46.180 Ashwini Sharma: And Daxter is orchestrating polyatomic pipelines, or…

757 01:13:46.520 ⇒ 01:13:49.150 Awaish Kumar: No, the extra is basically…

758 01:13:50.280 ⇒ 01:13:56.330 Awaish Kumar: We needed to run some… we needed to create some, basically, data pipelines for them.

759 01:13:56.670 ⇒ 01:14:01.190 Awaish Kumar: And for that, we… And they weren’t supported in… they weren’t supported.

760 01:14:01.190 ⇒ 01:14:13.410 Uttam Kumaran: in Polyatomic initially. The Polyatomic is just a 5chan alternative. There’s just way better support and cheaper, and they build… they’re building stuff for us faster, so…

761 01:14:13.610 ⇒ 01:14:23.600 Uttam Kumaran: But there’s times where they didn’t have things, or the scope was so small, like, it was, like, one endpoint that we needed some data from, so we just wrote the script to get it.

762 01:14:24.570 ⇒ 01:14:37.769 Uttam Kumaran: one… one thing I would tell you, now that you’re starting to own this, is you can make the call on asking Polytomic to build more. So if you’re like, hey, now we’re supporting, like, 3 or 4 endpoints from this source, now we have to maintain it.

763 01:14:38.130 ⇒ 01:14:41.279 Uttam Kumaran: We don’t want to do that, you can ask them to build it. I don’t care.

764 01:14:41.470 ⇒ 01:14:45.949 Uttam Kumaran: But we just requested them to build a GoHighLevel script.

765 01:14:46.050 ⇒ 01:14:49.800 Uttam Kumaran: connector and a, connector for UpFluence.

766 01:14:49.910 ⇒ 01:14:55.500 Uttam Kumaran: So, if you’re seeing that, hey, like, these scripts kind of fail, we want to transition, you could do that, you know?

767 01:14:57.400 ⇒ 01:14:58.040 Ashwini Sharma: Okay.

768 01:14:58.810 ⇒ 01:15:03.540 Ashwini Sharma: And that custom script is orchestrated through Daxter, right?

769 01:15:03.540 ⇒ 01:15:04.360 Awaish Kumar: I guess.

770 01:15:05.680 ⇒ 01:15:06.270 Ashwini Sharma: Okay.

771 01:15:12.050 ⇒ 01:15:19.349 Ashwini Sharma: Got it, okay. Let me finish that, I’ll try to finish that, Metaplane thing as soon as possible, and then I can pick up something else.

772 01:15:20.050 ⇒ 01:15:20.580 Uttam Kumaran: Okay.

773 01:15:28.520 ⇒ 01:15:29.230 Uttam Kumaran: Okay.

774 01:15:29.450 ⇒ 01:15:32.940 Uttam Kumaran: Let’s maybe talk about,

775 01:15:35.580 ⇒ 01:15:38.850 Uttam Kumaran: I don’t know who first? Maybe Amber? Wanna talk about…

776 01:15:38.980 ⇒ 01:15:52.360 Uttam Kumaran: two things, I think maybe we talk about insomnia first. I mentioned, and I was gonna send this, but I didn’t end up outside of my drafts, but I think we can also start to loop in Casey on analysis for insomnia.

777 01:15:52.540 ⇒ 01:15:55.220 Uttam Kumaran: I think you’ve set, like, a good foundation.

778 01:15:55.400 ⇒ 01:16:03.990 Uttam Kumaran: And I feel like you’re now starting to get pulled in a few different directions. We are bringing on one more analyst, sort of part-time.

779 01:16:04.180 ⇒ 01:16:06.350 Uttam Kumaran: As soon as she kind of signs stuff, but…

780 01:16:06.800 ⇒ 01:16:13.249 Uttam Kumaran: I think it’s probably a good opportunity for you to build some redundancy on Insomnia, and maybe you can loop in Casey to

781 01:16:13.410 ⇒ 01:16:19.580 Uttam Kumaran: Either source data, conduct some prime… initial analysis, but could be a good opportunity.

782 01:16:21.000 ⇒ 01:16:25.420 Amber Lin: Yeah, sounds good. What I… Then…

783 01:16:25.610 ⇒ 01:16:28.950 Amber Lin: How is the best way to get that started?

784 01:16:29.740 ⇒ 01:16:36.269 Uttam Kumaran: Like, do you have things on your plate that either need to get kicked off, or…

785 01:16:36.790 ⇒ 01:16:41.349 Uttam Kumaran: Are halfway done, that you feel like are in a good place to sort of loop someone in on?

786 01:16:43.650 ⇒ 01:16:55.550 Amber Lin: I am doing the extended analysis right now. I do need to do the opportunity sizing for the segments, so I can probably grab time crazy and

787 01:16:55.970 ⇒ 01:17:00.230 Amber Lin: tell him what I did so far, so that if new tasks come in, he can take the…

788 01:17:00.930 ⇒ 01:17:02.330 Uttam Kumaran: Okay. Yeah.

789 01:17:02.420 ⇒ 01:17:14.290 Uttam Kumaran: maybe, Rico, if you want to help schedule that for both of them, that would be great. And then, yeah, I think starting Monday, I want to start looping in Casey more on Insomnia. Similarly, Mustafa is also going to be…

790 01:17:14.290 ⇒ 01:17:23.379 Uttam Kumaran: setting up dbt for Honey Stinger, and probably as soon as, Amber, you get to a decent point on some analysis, you can start to loop them in on things.

791 01:17:23.600 ⇒ 01:17:32.090 Uttam Kumaran: Kind of, there’s two goals here. One, I think Casey and Mustafa, I want them to get exposure to the analysis side, but also just redundancy, like.

792 01:17:32.310 ⇒ 01:17:38.570 Uttam Kumaran: for some of these clients, like Insomnia, you’re now the primary developer, and…

793 01:17:38.870 ⇒ 01:17:52.460 Uttam Kumaran: like, you’re just never gonna be able to, like, go to bed at night. Things are gonna stop if you don’t do them, right? So, I want to constantly be looking at plot… trying to play the game of, like, transitioning, because it helps us build a muscle of

794 01:17:52.550 ⇒ 01:18:08.760 Uttam Kumaran: redundancy. Like, similarly, Ashwini is coming on Eden, now we’re seeing, like, okay, how hard is it for a new person to come on to a client? And I think, Ashwini, what I would love feedback from you on is, like, okay, what docs should we have created to streamline

795 01:18:09.210 ⇒ 01:18:14.239 Uttam Kumaran: the transition, right? Like, what would you have needed to help you

796 01:18:14.730 ⇒ 01:18:17.869 Uttam Kumaran: beat up your understanding of the client. So, like.

797 01:18:18.020 ⇒ 01:18:30.760 Uttam Kumaran: the more handoffs we do, and the more explaining of, like, what you’ve done, it actually will help you understand, and then just gets more people involved. So, I feel like if we’re okay on that for Insomnia and Honey Stinger, like, that’s… that’s kind of, like, what I hope the…

798 01:18:30.900 ⇒ 01:18:46.310 Uttam Kumaran: the plan. That’s… that’s how it goes. And Amber, you’ll still lead analysis, but you’ll just have support, basically. So if there’s… instead of being like, oh, I have to work on 3 at a time, if there’s a low-hanging fruit one, or if there’s something that’s more early, you can pass things off.

799 01:18:48.040 ⇒ 01:18:49.889 Amber Lin: Cool, yeah, sounds good.

800 01:18:50.650 ⇒ 01:18:58.269 Amber Lin: So for Honey Stinger… Yesterday, ugh, it took me so long, I probably should have…

801 01:18:58.710 ⇒ 01:19:00.640 Uttam Kumaran: Asked for help earlier.

802 01:19:00.890 ⇒ 01:19:10.289 Amber Lin: But it’s Amazon data, and then some inventory data. Also trying to run some statistical models, but I feel like the data we have

803 01:19:10.460 ⇒ 01:19:17.150 Amber Lin: Especially some data we didn’t backfill long enough, so the model is pretty shady.

804 01:19:17.250 ⇒ 01:19:26.390 Amber Lin: But hopefully, it tells a better insight now. An additional finding is that These spikes… occur…

805 01:19:27.470 ⇒ 01:19:46.529 Amber Lin: in this seems to occur in a cyclical nature, so, for example, one of their big spikes that happened, that they were totally unexpected, for the energy choose, they actually had a spike of a similar magnitude in September. So, if they were watching.

806 01:19:46.780 ⇒ 01:19:50.490 Amber Lin: Something that has… a PO spike before.

807 01:19:50.650 ⇒ 01:19:56.430 Amber Lin: They could have better anticipated it happening, because it has happened in the past.

808 01:19:57.590 ⇒ 01:20:01.390 Amber Lin: So that’s another insight that I can add.

809 01:20:03.770 ⇒ 01:20:12.469 Amber Lin: like, I think I’ve hit the wall, or hit the cap on what I can do on Amazon, because I just…

810 01:20:12.860 ⇒ 01:20:28.590 Amber Lin: I just don’t have a good enough understanding of how these things work, so probably at this point, I need to learn more, talk to you, talk to Robert, see what people usually approach this, and then I’ll probably be inspired to do more.

811 01:20:30.050 ⇒ 01:20:32.350 Uttam Kumaran: Cool. There’s still Walmart data.

812 01:20:32.790 ⇒ 01:20:34.099 Uttam Kumaran: So, if you feel like…

813 01:20:34.100 ⇒ 01:20:34.640 Amber Lin: Yeah, yeah, yeah.

814 01:20:34.640 ⇒ 01:20:37.500 Uttam Kumaran: You’re at a blocker, you could take a look at that before the meeting.

815 01:20:37.710 ⇒ 01:20:42.650 Uttam Kumaran: I feel like that’s, like, the… that would be a nice cherry on top for us to talk through, because Sam did find something.

816 01:20:42.650 ⇒ 01:20:44.629 Amber Lin: When’s the… when’s the meeting?

817 01:20:45.730 ⇒ 01:20:48.260 Uttam Kumaran: Let me… let me add you to that.

818 01:20:48.550 ⇒ 01:20:57.540 Uttam Kumaran: Okay. It’s, it is… At… 2 o’clock your time.

819 01:20:58.210 ⇒ 01:20:59.699 Amber Lin: Okay. Yeah, I’m free.

820 01:20:59.700 ⇒ 01:21:00.700 Uttam Kumaran: And…

821 01:21:01.160 ⇒ 01:21:06.389 Uttam Kumaran: It’s 6 AM Robert’s time, but he said he’ll be awake, so I’ll call him at, like, 5.30. I’ll call him at…

822 01:21:06.390 ⇒ 01:21:07.490 Amber Lin: Okay.

823 01:21:07.490 ⇒ 01:21:10.490 Uttam Kumaran: 30 has time to make sure he’s awake. Get up!

824 01:21:11.080 ⇒ 01:21:17.059 Uttam Kumaran: Get up. Alright, let me… I added you to this meeting.

825 01:21:17.760 ⇒ 01:21:18.400 Amber Lin: Cool.

826 01:21:18.860 ⇒ 01:21:22.989 Uttam Kumaran: Cool, and I think this is nice, like, kind of how we’re getting new clients.

827 01:21:23.590 ⇒ 01:21:28.530 Uttam Kumaran: Robert and I, or a mix of people, sort of set it up, and then we start to loop people in. I think that’s, like…

828 01:21:29.090 ⇒ 01:21:33.580 Uttam Kumaran: the way… We’re sort of, like, doing things these days, so…

829 01:21:34.940 ⇒ 01:21:39.380 Uttam Kumaran: Unless it’s, like, a huge client, but I think this… this pacing is good, so…

830 01:21:39.600 ⇒ 01:21:46.529 Uttam Kumaran: I looped you in there, great. Okay, so I feel good about both of those. Is there anything on insomnia, like, for today, really?

831 01:21:47.430 ⇒ 01:21:57.810 Amber Lin: Mostly, I’m just doing a before-after comparison of when their boxes order spiked. So far, it looks like

832 01:21:58.190 ⇒ 01:21:59.020 Amber Lin: the…

833 01:21:59.110 ⇒ 01:22:18.359 Amber Lin: average order size decreased, which makes sense, because they’re getting boxes. Also looking at conversion rates seems to be down, but also because, like, it’s a more recent event, so I’m not sure how much of it is because people haven’t had time to order yet, but it did… it’s a pretty…

834 01:22:19.030 ⇒ 01:22:22.059 Amber Lin: Obvious difference when it dipped.

835 01:22:22.420 ⇒ 01:22:31.979 Amber Lin: In June, when they started selling more boxes. So the conversion rates between the first order and second order is going down alongside the others.

836 01:22:32.740 ⇒ 01:22:33.280 Uttam Kumaran: Okay.

837 01:22:34.000 ⇒ 01:22:34.520 Amber Lin: Yeah.

838 01:22:37.780 ⇒ 01:22:39.959 Amber Lin: Are we sending a tag to them today?

839 01:22:41.740 ⇒ 01:22:42.290 Uttam Kumaran: gorgeous.

840 01:22:42.290 ⇒ 01:22:45.260 Amber Lin: I can probably just share the insights in the channel.

841 01:22:45.260 ⇒ 01:22:47.440 Uttam Kumaran: Yeah, if you can sh- if you can share the insights…

842 01:22:47.700 ⇒ 01:22:52.310 Uttam Kumaran: And then, I… yeah, you can… if you share the insights, that would be great.

843 01:22:52.310 ⇒ 01:22:52.870 Amber Lin: Cool.

844 01:22:53.540 ⇒ 01:22:54.100 Uttam Kumaran: Yeah.

845 01:23:04.560 ⇒ 01:23:07.160 Uttam Kumaran: Okay, let’s talk about,

846 01:23:07.330 ⇒ 01:23:14.249 Uttam Kumaran: Eden Zuron, so maybe we can… Yeah, I would love to just maybe have you briefly walk me through your…

847 01:23:14.540 ⇒ 01:23:21.209 Uttam Kumaran: your chart, and then I can… I’m gonna be, like, in planning mode rest of the day, so I will…

848 01:23:21.670 ⇒ 01:23:26.000 Uttam Kumaran: can just take a look at stuff. Let me… I’ll just pull it up on my side, give me one second.

849 01:23:26.000 ⇒ 01:23:26.750 Zoran Selinger: Okay.

850 01:23:29.030 ⇒ 01:23:30.719 Uttam Kumaran: And thanks, guys, for, like…

851 01:23:30.840 ⇒ 01:23:39.199 Uttam Kumaran: adopting this Gantt charts, I feel like these are huge. I mean, it’s, I don’t know what you guys think, but it’s giving me a lot of visibility into things, so…

852 01:23:39.250 ⇒ 01:23:48.429 Zoran Selinger: Gantt is generally fine. I haven’t… I don’t think I’ve… I’ve made a Gantt chart since college.

853 01:23:48.430 ⇒ 01:23:56.309 Uttam Kumaran: No, no, I also don’t think I made one in the… in the wild in a… I did one at one of my jobs.

854 01:23:56.860 ⇒ 01:24:02.279 Uttam Kumaran: But I was, like, being, like, a know-it-all, like, I probably didn’t need to do that.

855 01:24:04.510 ⇒ 01:24:08.760 Uttam Kumaran: But, like, I’m telling you, the clients love Gantt chart.

856 01:24:08.760 ⇒ 01:24:20.820 Zoran Selinger: Okay. I mean, it’s not… it’s not a problem, and this is… it’s… this… this one’s not, like, not too many features or anything like that. We just keep it simple, just a few dependencies and…

857 01:24:20.990 ⇒ 01:24:26.880 Zoran Selinger: It looks, it looks good. It’s a good one.

858 01:24:27.710 ⇒ 01:24:36.590 Uttam Kumaran: Yeah, and like, think about it from the executives, like, they don’t have any clue about, like, what we’re doing, or how we’re doing it, so as many things as we can provide to them.

859 01:24:36.820 ⇒ 01:24:39.800 Uttam Kumaran: To, like, sort of show that we’re organized.

860 01:24:40.170 ⇒ 01:24:45.630 Uttam Kumaran: Is, like, you know, so… Okay, great. So yeah, maybe you want to just, walk me through things.

861 01:24:46.240 ⇒ 01:25:05.799 Zoran Selinger: Yeah, so for NordBeam, we have to do the implementation audit, we can do that basically right away, and we need to perform UTM audit. So the UTM audit is important for our non-integrated channels as well.

862 01:25:06.870 ⇒ 01:25:14.429 Zoran Selinger: because we will use UTMs to define those non-integrated channels.

863 01:25:14.550 ⇒ 01:25:15.530 Zoran Selinger: And is this…

864 01:25:15.530 ⇒ 01:25:16.180 Uttam Kumaran: So…

865 01:25:16.290 ⇒ 01:25:21.809 Uttam Kumaran: Is this… is this mainly you’re looking at traffic that’s coming in without UTMs on it? Like, how do you do this automated?

866 01:25:22.920 ⇒ 01:25:40.640 Zoran Selinger: So, we want to look at… we want to do this with our edge layer tables, so our… our… our kind of session… session table on edge, because we have all the UTMs in there, and we want to catch anything non-standard that we haven’t agreed on.

867 01:25:40.780 ⇒ 01:25:44.580 Zoran Selinger: So, a part of this task is also going to be me checking

868 01:25:44.680 ⇒ 01:25:58.089 Zoran Selinger: Do we actually, do we actually, abide by what we previously agreed, our, like, our UTM template, right? So, we should do that. This should be a recurring task, though.

869 01:25:59.410 ⇒ 01:26:02.959 Zoran Selinger: We should do this at least Twice a month.

870 01:26:03.940 ⇒ 01:26:04.560 Uttam Kumaran: Okay.

871 01:26:05.080 ⇒ 01:26:23.200 Zoran Selinger: Yeah, so the UTM check, if we can automate it, we should do it every week, and basically, I envision just having some kind of email with a list of non-standard UTMs that are outside of our template, our UTM template, and we just talk to

872 01:26:23.400 ⇒ 01:26:35.100 Zoran Selinger: to Ethan about, okay, guys, on this channel, we see something non-standard, either wrong capitalization on something that’s… that we… or… or a UTM that we never discussed before.

873 01:26:35.520 ⇒ 01:26:40.370 Zoran Selinger: But that should be, you know, a short discussion, maybe every week.

874 01:26:42.120 ⇒ 01:26:55.270 Zoran Selinger: Just so they can clean up, because that always happens, right? We can agree on… on a UTM template. Adrift from that standard will always happen. Not constantly, but sometimes it will happen.

875 01:26:57.850 ⇒ 01:26:58.680 Zoran Selinger: Yeah.

876 01:26:59.480 ⇒ 01:27:00.660 Zoran Selinger: So, yeah…

877 01:27:00.660 ⇒ 01:27:03.159 Uttam Kumaran: Maybe this is something we can do in dbt.

878 01:27:03.400 ⇒ 01:27:06.670 Uttam Kumaran: But yeah, totally, we should automate this.

879 01:27:06.670 ⇒ 01:27:07.190 Zoran Selinger: Yep.

880 01:27:07.730 ⇒ 01:27:09.280 Uttam Kumaran: identifying…

881 01:27:12.880 ⇒ 01:27:21.369 Zoran Selinger: We do have some non-standard already, even… even some, some, some are… look kind of really weird.

882 01:27:21.590 ⇒ 01:27:22.770 Zoran Selinger: So we want.

883 01:27:22.770 ⇒ 01:27:25.370 Uttam Kumaran: Can you, can you link some of, can you link some of those here?

884 01:27:26.130 ⇒ 01:27:29.560 Zoran Selinger: Orion has… .

885 01:27:29.560 ⇒ 01:27:33.949 Uttam Kumaran: Or even if there’s a Slack conversation, you can… put it here.

886 01:27:34.660 ⇒ 01:27:37.439 Zoran Selinger: Yeah, okay, cool, cool. I can do that.

887 01:27:44.930 ⇒ 01:27:46.360 Uttam Kumaran: Great, okay, perfect.

888 01:27:46.440 ⇒ 01:27:50.899 Zoran Selinger: Cool. So then, we have our edge layer.

889 01:27:51.000 ⇒ 01:27:52.810 Zoran Selinger: We need to add.

890 01:27:52.810 ⇒ 01:27:54.759 Uttam Kumaran: What is this? What is the data drift?

891 01:27:57.130 ⇒ 01:28:10.439 Zoran Selinger: Oh, yeah, so what we just did, the analysis we did this week, to conclude, okay, we have… we’re missing only 1% of transactions that we have on the edge, then we’re missing some of,

892 01:28:10.590 ⇒ 01:28:12.579 Zoran Selinger: some, touchpoints.

893 01:28:12.940 ⇒ 01:28:20.540 Zoran Selinger: In NordBeam, compared to what we have on Edge. So that… we should do it quarterly, I think, is… is good enough.

894 01:28:23.280 ⇒ 01:28:27.440 Zoran Selinger: We… that’s a 3-4 hours of work.

895 01:28:29.580 ⇒ 01:28:31.010 Zoran Selinger: But we should do it.

896 01:28:31.400 ⇒ 01:28:32.770 Zoran Selinger: Probably quarterly.

897 01:28:34.490 ⇒ 01:28:35.080 Uttam Kumaran: Okay.

898 01:28:35.770 ⇒ 01:28:43.949 Zoran Selinger: So next, next audit would be in, in, in February, basically. That’s why I put the date there.

899 01:28:44.310 ⇒ 01:28:45.749 Uttam Kumaran: Okay, okay, great, great, great.

900 01:28:45.830 ⇒ 01:28:46.780 Zoran Selinger: Yeah, cool.

901 01:28:47.350 ⇒ 01:28:48.030 Uttam Kumaran: Great.

902 01:28:48.870 ⇒ 01:28:53.589 Uttam Kumaran: And then UTM Audit, yeah, this is gonna be… the… okay, so this one…

903 01:28:53.590 ⇒ 01:28:58.869 Zoran Selinger: It’s a recurring one, I don’t know if we can put recurring tasks here in the chart.

904 01:28:59.210 ⇒ 01:29:02.830 Uttam Kumaran: Yeah, I’ll take a look at, like, how we can do that, or… yeah.

905 01:29:03.130 ⇒ 01:29:03.560 Zoran Selinger: Okay.

906 01:29:03.560 ⇒ 01:29:04.460 Uttam Kumaran: Okay, great.

907 01:29:04.460 ⇒ 01:29:05.000 Zoran Selinger: Cool.

908 01:29:05.380 ⇒ 01:29:11.189 Zoran Selinger: Then, yeah, we have our, we have a few improvements for our, Edge.

909 01:29:11.570 ⇒ 01:29:13.399 Zoran Selinger: edge function.

910 01:29:14.100 ⇒ 01:29:24.959 Zoran Selinger: So the big ones are this update from Basque, and we want to add a few more identifiers as well. So that’s all ticketed in linear.

911 01:29:25.380 ⇒ 01:29:33.659 Zoran Selinger: And then we have this nice-to-have of limiting the number of invocations of that worker.

912 01:29:33.820 ⇒ 01:29:38.409 Zoran Selinger: To, you know, only session starts and thank you page visits.

913 01:29:38.650 ⇒ 01:29:44.389 Zoran Selinger: This is just a nice-to-have, really.

914 01:29:44.800 ⇒ 01:29:51.910 Zoran Selinger: But I would like to do it, because we will save them some money. I mean, they’re paying very little anyway, right now.

915 01:29:52.520 ⇒ 01:30:02.949 Zoran Selinger: those 6… I don’t know, that’s 60, 70 million requests a month, that still, like, cost them 20 bucks. It’s nothing, really.

916 01:30:03.090 ⇒ 01:30:16.550 Zoran Selinger: But we… I mean, still, there’s a lot of unnecessary firing, so I would just like to do it. Obviously, this does not take that long. I’ll just… we’ll need, it’ll probably,

917 01:30:17.340 ⇒ 01:30:19.940 Zoran Selinger: Be, you know, over a week.

918 01:30:20.160 ⇒ 01:30:32.990 Zoran Selinger: of, you know, discussing, maybe going back and forth with Ryan, a little bit, investigating how we can actually do it. We’ll have to use some transform rules, unfortunately, here, which are dangerous

919 01:30:33.280 ⇒ 01:30:44.700 Zoran Selinger: So this is why we are kind of postponing this, and kind of afraid to deal with this particular task. It is a dangerous one. They already have transform rules.

920 01:30:44.910 ⇒ 01:30:53.750 Zoran Selinger: And they do interact with each other, and the sequence is important, so we have to be extremely careful with this. The problem is that

921 01:30:54.090 ⇒ 01:31:04.970 Zoran Selinger: Like, routing rules for workers are very, very simple, and we cannot… We cannot,

922 01:31:05.800 ⇒ 01:31:12.470 Zoran Selinger: par them based on query parameters. So we have to internally rewrite URLs.

923 01:31:13.790 ⇒ 01:31:23.369 Zoran Selinger: When there are credit parameters, and yes, and then trigger our… our workers based on that virtual route, and… which is dangerous, so…

924 01:31:23.370 ⇒ 01:31:23.979 Uttam Kumaran: Yeah, yeah, yeah.

925 01:31:23.980 ⇒ 01:31:27.619 Zoran Selinger: to, yeah, be very, very careful. This is why I put it, like.

926 01:31:27.770 ⇒ 01:31:31.919 Zoran Selinger: It’ll take us a week to be sure of what we’re doing there.

927 01:31:32.090 ⇒ 01:31:35.150 Uttam Kumaran: Okay. So we don’t bring anything down, yeah.

928 01:31:35.850 ⇒ 01:31:52.100 Zoran Selinger: Okay. So for Meta, yeah, so we wanna… so Avesh already finished modeling there. There might be just a few, few tweaks there, in terms of… when I review UTMs, there might be just a few tweaks.

929 01:31:52.450 ⇒ 01:31:55.299 Zoran Selinger: the modeling is done.

930 01:31:56.630 ⇒ 01:32:07.229 Zoran Selinger: so there’s a task missing here. I will try to configure, the actual API requests from that table into,

931 01:32:07.980 ⇒ 01:32:09.420 Zoran Selinger: conversion API.

932 01:32:09.580 ⇒ 01:32:10.230 Zoran Selinger: Permit?

933 01:32:10.230 ⇒ 01:32:11.099 Uttam Kumaran: Yeah, yeah, yeah.

934 01:32:11.100 ⇒ 01:32:20.980 Zoran Selinger: So I… I will ask for advice on how do… how you typically do this, what do you want to use? Avesh mentioned a segment.

935 01:32:21.600 ⇒ 01:32:21.930 Uttam Kumaran: Yeah.

936 01:32:21.930 ⇒ 01:32:24.990 Zoran Selinger: I’m not using that way. So I have the table.

937 01:32:25.470 ⇒ 01:32:37.869 Uttam Kumaran: Yeah, if you wanna… even if you wanna… if you’re thinking about this, like, I guess my point here is join one of the engineering calls next week, like the earlier call, and we could discuss. Yeah.

938 01:32:37.870 ⇒ 01:32:39.890 Zoran Selinger: Oh, right, okay, okay. Yeah.

939 01:32:39.890 ⇒ 01:32:45.859 Uttam Kumaran: So there’s an engineering call right before this stand-up, so you can join that, and if anything engineering, we can talk there, for sure.

940 01:32:46.420 ⇒ 01:32:47.060 Zoran Selinger: Okay.

941 01:32:47.600 ⇒ 01:32:49.340 Zoran Selinger: Cool. Boom.

942 01:32:49.500 ⇒ 01:32:53.649 Zoran Selinger: So, we have this catalyst, we have to, again.

943 01:32:54.230 ⇒ 01:33:03.149 Zoran Selinger: starts working on… on… So, Vaish, I… I would like that to be your next task.

944 01:33:03.660 ⇒ 01:33:11.420 Zoran Selinger: When it comes to marketing, that’s definitely your next marketing task, or whoever’s helping, there as well.

945 01:33:11.810 ⇒ 01:33:15.259 Zoran Selinger: So we need to change the model.

946 01:33:15.260 ⇒ 01:33:15.910 Awaish Kumar: Yeah.

947 01:33:15.910 ⇒ 01:33:17.999 Zoran Selinger: We’re going back to 4th.

948 01:33:18.080 ⇒ 01:33:18.830 Awaish Kumar: Yeah.

949 01:33:19.000 ⇒ 01:33:22.909 Awaish Kumar: I will create a table, and I will assign to Ashwani.

950 01:33:23.650 ⇒ 01:33:26.730 Zoran Selinger: Okay, okay, excellent, thank you, thank you.

951 01:33:27.130 ⇒ 01:33:32.960 Zoran Selinger: So yeah, that’s just one thing there. I think, if…

952 01:33:33.200 ⇒ 01:33:41.169 Zoran Selinger: They will probably, ask for… to go back, retroactively, and,

953 01:33:42.200 ⇒ 01:33:45.369 Zoran Selinger: You know, we’ve had this kind of strict

954 01:33:45.910 ⇒ 01:33:54.340 Zoran Selinger: crediting policy for them for the last, what, 2 weeks? They will probably ask us to manually go back and give them more credit based on…

955 01:33:54.990 ⇒ 01:33:59.429 Zoran Selinger: You know, this new… new old rule that we agreed on.

956 01:33:59.920 ⇒ 01:34:10.040 Zoran Selinger: So, basically, we are going to credit them with, with a conversion for anything, for any interaction that they were, touchpoint.

957 01:34:11.780 ⇒ 01:34:13.370 Zoran Selinger: From what we had before.

958 01:34:14.120 ⇒ 01:34:18.690 Zoran Selinger: Regardless of the intake they go through. So, yeah.

959 01:34:18.750 ⇒ 01:34:26.419 Awaish Kumar: Okay, so we are disregarding the intake slugs, right?

960 01:34:26.440 ⇒ 01:34:29.380 Zoran Selinger: Yes, slides will not be important in this.

961 01:34:29.380 ⇒ 01:34:36.619 Awaish Kumar: Okay. And also, we… we are going to use any touch in the last 14 days, right?

962 01:34:37.200 ⇒ 01:34:49.350 Zoran Selinger: Let me confirm that. Is it any touch, or it has to be the first touch? But we will, yes, we will have to look up all the, all the, touch points. But let me confirm that.

963 01:34:49.350 ⇒ 01:34:50.500 Awaish Kumar: Yeah.

964 01:34:51.030 ⇒ 01:34:55.820 Awaish Kumar: Yeah, but… what I remember from last meeting with Mitesh, it was, like.

965 01:34:55.820 ⇒ 01:34:58.970 Zoran Selinger: Any touching last 14 days, but you can confirm here.

966 01:34:59.160 ⇒ 01:35:04.540 Zoran Selinger: Yes, yes, that’s… that’s exactly what I remember, but I just wanna… I wanna be sure there.

967 01:35:06.300 ⇒ 01:35:16.469 Zoran Selinger: So affluence, I… actually, I haven’t had a chance. I have no… I had no conversation about affluence yet.

968 01:35:16.940 ⇒ 01:35:21.360 Zoran Selinger: I think, I think, Henry knows what to do there.

969 01:35:22.140 ⇒ 01:35:23.020 Uttam Kumaran: Okay, yep.

970 01:35:24.290 ⇒ 01:35:28.480 Zoran Selinger: So… Yeah, so we’ll have to do something like that.

971 01:35:28.480 ⇒ 01:35:28.930 Uttam Kumaran: What happened?

972 01:35:29.280 ⇒ 01:35:31.399 Uttam Kumaran: So, let’s say we get the data.

973 01:35:31.760 ⇒ 01:35:35.460 Uttam Kumaran: what happens after, like, the data is in BigQuery?

974 01:35:37.540 ⇒ 01:35:38.200 Zoran Selinger: Yeah, so we…

975 01:35:38.200 ⇒ 01:35:39.319 Uttam Kumaran: sitting on it? Yeah.

976 01:35:39.960 ⇒ 01:35:52.859 Zoran Selinger: Yeah, we have to figure out how to get it in the platform. But I was just… I was… I just asked Vanessa, was that Vanessa Gomez? They’re dealing with that channel, I guess, internally?

977 01:35:53.230 ⇒ 01:36:10.270 Zoran Selinger: how do we identify exactly the… the traffic? I still… like I said, I haven’t had any conversation about affluence yet, so I don’t even know what UTMs look like for that channel or anything like that. So I’m not ready to work on this at all.

978 01:36:10.840 ⇒ 01:36:12.180 Uttam Kumaran: Yeah, so, so…

979 01:36:12.330 ⇒ 01:36:18.750 Uttam Kumaran: we’re working on this. So you just… but I guess what I need to know is, like, after the data is in BigQuery.

980 01:36:18.870 ⇒ 01:36:22.599 Uttam Kumaran: like, what are your steps? Is there gonna… there’s basically, like, a discovery?

981 01:36:23.210 ⇒ 01:36:42.030 Awaish Kumar: I think there are two things here. What we have asked right now from… from, sorry, the Polytomic is ETL. So, we are going to receive the spent data from their, their apprentice platform to the BigQuery, and that will be used in the…

982 01:36:42.030 ⇒ 01:36:54.519 Awaish Kumar: reporting. But what maybe Zoran is talking about is more like reverse ETL, where we have to push the data to upfronts, and maybe, like.

983 01:36:54.700 ⇒ 01:36:57.920 Awaish Kumar: Like, for that attribution of the orders.

984 01:36:58.690 ⇒ 01:36:59.290 Uttam Kumaran: Okay.

985 01:36:59.290 ⇒ 01:37:04.090 Zoran Selinger: Exactly. Edge… so basically, anything edge is a versatile.

986 01:37:05.020 ⇒ 01:37:06.020 Uttam Kumaran: Yeah, yeah, yeah, yeah.

987 01:37:06.020 ⇒ 01:37:09.670 Zoran Selinger: Yeah, everything here is reverse ETL, that’s what I’m talking about, yeah.

988 01:37:12.110 ⇒ 01:37:22.429 Awaish Kumar: Yeah, so if you have any requirements, or if you can gather anything, let us know. We are going to ask Polytomic regarding your CTL connector as well.

989 01:37:24.140 ⇒ 01:37:32.609 Zoran Selinger: So I think logic, surveys logic here will be pretty similar for all the channels, which,

990 01:37:32.740 ⇒ 01:37:41.249 Zoran Selinger: which makes sense for almost all the channels. These are all basically top funnel, first, like.

991 01:37:41.640 ⇒ 01:37:45.680 Zoran Selinger: generating first-time customers, all that, so I think,

992 01:37:45.950 ⇒ 01:37:49.869 Zoran Selinger: Crediting logic will be pretty similar for all of them.

993 01:37:51.220 ⇒ 01:38:08.490 Zoran Selinger: So once we… we already kind of, went through… through a few exercises with Catalysts, so, yeah, we’re ready to deal with that. It’s just gonna be… you’re gonna look up different UTMs, UTM mediums, and… and…

994 01:38:08.760 ⇒ 01:38:10.010 Zoran Selinger: That’s gonna be that.

995 01:38:10.010 ⇒ 01:38:11.170 Awaish Kumar: Have a look at…

996 01:38:11.170 ⇒ 01:38:17.069 Zoran Selinger: I’ll confirm Logic separately for every channel, of course, before we go into that work.

997 01:38:19.300 ⇒ 01:38:20.520 Zoran Selinger: Yeah, so…

998 01:38:20.520 ⇒ 01:38:27.519 Awaish Kumar: I’ll… I will confirm if, like, if there is any API endpoints available for Diverse CTL, or…

999 01:38:27.700 ⇒ 01:38:30.149 Awaish Kumar: If we… if we need access to those.

1000 01:38:31.680 ⇒ 01:38:32.370 Zoran Selinger: Okay.

1001 01:38:33.330 ⇒ 01:38:38.889 Zoran Selinger: Yeah, and… and Utan, last thing is… is TikTok. This is,

1002 01:38:39.280 ⇒ 01:38:50.540 Zoran Selinger: This is… like, if we do… if we manage to do a meta, and we will, this one should be an easy one, to recreate for.

1003 01:38:52.680 ⇒ 01:38:59.069 Zoran Selinger: they would be really happy with us if we get… if we configure that one in… in December as well.

1004 01:39:01.200 ⇒ 01:39:06.319 Zoran Selinger: Okay, okay. So yeah, TikTok, but can’t you get the access now?

1005 01:39:08.890 ⇒ 01:39:10.310 Zoran Selinger: Yeah, yeah.

1006 01:39:10.580 ⇒ 01:39:12.789 Uttam Kumaran: So just go ask for it, yeah, just ask for it.

1007 01:39:16.150 ⇒ 01:39:24.680 Uttam Kumaran: Sorry, I mean, not to make it… not to, like… you don’t… I guess what I’m saying is, like… so there’s… there’s two things, I think, since we’re at the end of this. One is, like…

1008 01:39:24.890 ⇒ 01:39:32.619 Uttam Kumaran: and this is sort of the feedback I gave Henry, is on the Gantt chart, anything that’s less than a week, I’m nervous about, because…

1009 01:39:33.400 ⇒ 01:39:47.940 Uttam Kumaran: what’s 3 days versus 2 days versus 5 days, right? So, if it’s less than 3 days, then… and it’s truly, like, okay, it’s gonna take a couple hours, fine. But if it’s, like, 3 days, then I would just round up to a week, because…

1010 01:39:47.940 ⇒ 01:39:48.430 Zoran Selinger: Okay.

1011 01:39:48.430 ⇒ 01:39:50.419 Uttam Kumaran: You know, anything can go wrong, right? So…

1012 01:39:50.420 ⇒ 01:39:53.470 Zoran Selinger: Of course, I mean, I agree with you.

1013 01:39:53.470 ⇒ 01:39:56.300 Uttam Kumaran: What I’m not looking for is,

1014 01:39:56.620 ⇒ 01:39:59.800 Uttam Kumaran: Aspiration, I’m looking for accuracy, you know?

1015 01:40:00.250 ⇒ 01:40:05.070 Uttam Kumaran: And we want to… we want to under-promise, over-deliver, so…

1016 01:40:05.310 ⇒ 01:40:11.310 Uttam Kumaran: even if the chart gets bigger, I don’t… I don’t mind, but I want to start in, like, what is the worst case scenario?

1017 01:40:11.750 ⇒ 01:40:12.540 Uttam Kumaran: You know?

1018 01:40:13.000 ⇒ 01:40:13.720 Uttam Kumaran: Option.

1019 01:40:13.720 ⇒ 01:40:21.930 Zoran Selinger: I mean, if this is… if you… if your full intention is to… to show this to the client, then of course.

1020 01:40:22.320 ⇒ 01:40:23.250 Zoran Selinger: I’m…

1021 01:40:23.250 ⇒ 01:40:25.080 Uttam Kumaran: No, that’s my intention, dude, yeah.

1022 01:40:25.380 ⇒ 01:40:30.390 Zoran Selinger: Okay, cool, I understand. I can adjust that.

1023 01:40:30.790 ⇒ 01:40:40.790 Uttam Kumaran: No, that’s fine, and yeah, yeah, so that’s, like, that’s the one thing. So yeah, our goal is to show this, like, I want them to be bought in, I want them to be… I want them to say, like.

1024 01:40:40.790 ⇒ 01:40:57.400 Uttam Kumaran: what is this upluence thing? Oh, okay, this, and then, like, yeah, actually, push that out. TikTok is more important. Like, that’s what we want them to give us priority on. I want them to see this, this, this like this. But yes, like, I would not promise anything under a week SLA for, like, net new items.

1025 01:40:57.520 ⇒ 01:40:59.249 Uttam Kumaran: Some of these, of course, are, like.

1026 01:40:59.440 ⇒ 01:41:04.480 Uttam Kumaran: meetings, right? So for get TikTok access, I would turn this into, like.

1027 01:41:04.800 ⇒ 01:41:12.189 Uttam Kumaran: more of a milestone. Like, this is just, like, a one-day thing. And yeah, like, ideally, if you can do that now.

1028 01:41:12.390 ⇒ 01:41:15.330 Uttam Kumaran: for example, if you get TikTok access, and then…

1029 01:41:15.540 ⇒ 01:41:19.750 Uttam Kumaran: could hand that off to a Shweeny, for example, to say, hey, figure out how to land this in the…

1030 01:41:19.930 ⇒ 01:41:23.289 Uttam Kumaran: thing. Those can all happen without your…

1031 01:41:23.730 ⇒ 01:41:32.820 Uttam Kumaran: like, you don’t need to spend any time there until the data’s landed, right? Because this is… this is, like, what you’re working on. These two, the data engineers will…

1032 01:41:32.920 ⇒ 01:41:39.709 Uttam Kumaran: Full handle, so… But… Again, that’s… I think that’s where you can kind of do some of the sequencing.

1033 01:41:40.230 ⇒ 01:41:47.599 Uttam Kumaran: Similarly on these, like, I think for any new source, we have to set up a connector, configure the data model, but those…

1034 01:41:48.310 ⇒ 01:41:53.880 Uttam Kumaran: I think as long as those are there, I… you know, we can work in the engineering sync to sort of plan it out.

1035 01:41:54.130 ⇒ 01:41:57.930 Uttam Kumaran: And so that’s… that’s great.

1036 01:41:58.110 ⇒ 01:42:04.320 Uttam Kumaran: I think… yeah, I feel pretty good on the fact that, like, there are… blockers here.

1037 01:42:04.660 ⇒ 01:42:10.440 Uttam Kumaran: I think… My only push is gonna be, okay, what happens after December?

1038 01:42:11.590 ⇒ 01:42:13.560 Zoran Selinger: Yeah, I know, I…

1039 01:42:13.560 ⇒ 01:42:18.090 Uttam Kumaran: So this is where, like, if you’re like, hey, we need… maybe we should spend some time

1040 01:42:18.510 ⇒ 01:42:30.650 Uttam Kumaran: discussing, like, what is the Q1 roadmap, we can do that too, but of course, this is, like, we want to at least have a… have an insight into between now and the end of the year, so… yeah.

1041 01:42:31.990 ⇒ 01:42:35.110 Zoran Selinger: Yeah, so it’s… it’s gonna depend on…

1042 01:42:35.760 ⇒ 01:42:40.120 Zoran Selinger: So, they are about to launch, launch new…

1043 01:42:40.270 ⇒ 01:42:49.049 Zoran Selinger: new channels. They really wanna go into meta properly and all that, so it’s really gonna depend on the channels that they wanna…

1044 01:42:49.450 ⇒ 01:42:51.060 Zoran Selinger: They wanna activate.

1045 01:42:53.040 ⇒ 01:42:59.679 Zoran Selinger: So, yeah, we’ll see, we’ll see. We can… when do we need to have an idea on the… on the Q1?

1046 01:43:00.850 ⇒ 01:43:02.359 Uttam Kumaran: Anytime.

1047 01:43:02.360 ⇒ 01:43:02.690 Zoran Selinger: on the.

1048 01:43:02.690 ⇒ 01:43:13.289 Uttam Kumaran: I would… I would prefer that even if you, Even if we, what’s it called?

1049 01:43:14.290 ⇒ 01:43:18.800 Uttam Kumaran: even if you’re like, I… I have even, like, a couple things we want to do, throw them in here.

1050 01:43:19.090 ⇒ 01:43:25.310 Uttam Kumaran: But by the end… by the mid-next… by mid-next month, I would like to have a Q1 understanding.

1051 01:43:25.510 ⇒ 01:43:26.110 Zoran Selinger: Okay.

1052 01:43:26.380 ⇒ 01:43:26.870 Zoran Selinger: Okay.

1053 01:43:26.870 ⇒ 01:43:33.140 Uttam Kumaran: So, I think Robert will be meeting again with leadership

1054 01:43:34.750 ⇒ 01:43:42.800 Uttam Kumaran: on Wednesday, so the next one… by the next one, I think we can have a view of Q1. That would be ideal.

1055 01:43:43.780 ⇒ 01:43:44.330 Zoran Selinger: Okay.

1056 01:43:47.130 ⇒ 01:43:55.399 Uttam Kumaran: Cool, okay, I feel pretty good about this. I think I’m learning a little bit more about how, like, yeah, everything is basically reverse TL back into…

1057 01:43:55.790 ⇒ 01:44:00.300 Uttam Kumaran: things. I think…

1058 01:44:00.900 ⇒ 01:44:06.249 Uttam Kumaran: Yeah, I feel… so I feel overall pretty good. I think the only things I want to add here are, like.

1059 01:44:06.340 ⇒ 01:44:25.570 Uttam Kumaran: how does… how do we end up demonstrating the value of a lot of this work in the weekly… in the bi-weekly meetings? So we should think also a little bit about, like, what KPIs are we driving towards? For example, we should look at out-of-compliance UTMs, that’s a great KPI, right? Or et cetera, et cetera. So we can talk through that as well.

1060 01:44:26.080 ⇒ 01:44:26.660 Zoran Selinger: Okay.

1061 01:44:29.740 ⇒ 01:44:33.510 Uttam Kumaran: Cool. Okay, great, I feel good about this.

1062 01:44:33.890 ⇒ 01:44:35.510 Uttam Kumaran: What else?

1063 01:44:36.160 ⇒ 01:44:42.550 Uttam Kumaran: We talked about… Eating stuff…

1064 01:44:44.000 ⇒ 01:44:48.089 Uttam Kumaran: Oh, Robert, you’re on. So we talked a little bit about, Honey Stinger.

1065 01:44:48.230 ⇒ 01:44:50.869 Uttam Kumaran: I think Amber’s in a good place with…

1066 01:44:51.930 ⇒ 01:44:56.459 Uttam Kumaran: stuff for you to review for that meeting, and then I said, as a bonus, take a look at Walmart stuff.

1067 01:44:58.660 ⇒ 01:45:14.800 Robert Tseng: Yeah, I reviewed her deck. I think I have enough to talk to him about. I would like to be able to give him updates on, like, new data that’s landed, so I can just tell him, okay, look, we ran all the way through on Amazon, we gotta look at Walmart, we gotta look at Shopify, and then we’re gonna be able to

1068 01:45:14.900 ⇒ 01:45:20.859 Robert Tseng: maybe fill in some of the gaps there. Like, I guess, just wondering if we have something like that to share with him.

1069 01:45:21.340 ⇒ 01:45:32.890 Uttam Kumaran: Yeah, so… Maybe what we can do in part of this deck is… start…

1070 01:45:33.280 ⇒ 01:45:35.510 Uttam Kumaran: Can I start the deck with, like…

1071 01:45:37.790 ⇒ 01:45:41.510 Uttam Kumaran: Some insight into, like, the data warehouse to date?

1072 01:45:42.080 ⇒ 01:45:42.660 Robert Tseng: Yeah.

1073 01:45:42.660 ⇒ 01:45:44.430 Uttam Kumaran: Or like, okay, okay, okay, great.

1074 01:45:45.500 ⇒ 01:45:47.360 Uttam Kumaran: So, maybe, Sam, I…

1075 01:45:47.360 ⇒ 01:45:54.840 Robert Tseng: Demi is their Amazon person, their Acosta person, I’m just gonna try to use it as a time to learn more things, to get some more leads.

1076 01:45:55.310 ⇒ 01:45:59.560 Uttam Kumaran: She has some other data, by the way. She has some retail data and stuff, so yeah, I…

1077 01:45:59.560 ⇒ 01:46:02.209 Robert Tseng: Were you able to open that spreadsheet? Because I still can’t.

1078 01:46:02.390 ⇒ 01:46:08.110 Uttam Kumaran: I didn’t get any… I didn’t… I didn’t even get any… actually, I mean, I’ll check, I feel like I didn’t get anything.

1079 01:46:08.650 ⇒ 01:46:13.179 Robert Tseng: Okay. I still think we can, so I don’t know, I guess, what data you’re referring to.

1080 01:46:13.740 ⇒ 01:46:17.830 Uttam Kumaran: No, no, no, she… so, in that… yeah, in… in that email.

1081 01:46:17.930 ⇒ 01:46:21.560 Uttam Kumaran: He said, hey, you’re giving us some data, like.

1082 01:46:21.730 ⇒ 01:46:25.560 Uttam Kumaran: Order history, brand page analytics, monthly, weekly trackers.

1083 01:46:25.750 ⇒ 01:46:29.129 Uttam Kumaran: I said, yeah, give me everything you have, basically.

1084 01:46:29.580 ⇒ 01:46:33.330 Uttam Kumaran: And… and then I don’t… I don’t have, like…

1085 01:46:33.670 ⇒ 01:46:36.620 Uttam Kumaran: that SharePoint URL, like, I don’t know what… I guess I…

1086 01:46:37.090 ⇒ 01:46:39.290 Uttam Kumaran: it didn’t come to me, or… I don’t know.

1087 01:46:40.700 ⇒ 01:46:42.229 Uttam Kumaran: This is my email.

1088 01:46:43.040 ⇒ 01:46:43.560 Robert Tseng: Yeah.

1089 01:46:43.560 ⇒ 01:46:46.569 Uttam Kumaran: I saw your reply, but I don’t have any attachments.

1090 01:46:48.410 ⇒ 01:46:55.440 Robert Tseng: Yeah, the attachment came in through a different email from Byron, which I feel like you’re CC’d on, but it’s fine. Like, I don’t think they really responded to it anyway.

1091 01:46:55.880 ⇒ 01:47:02.430 Uttam Kumaran: Send… yeah, send… send it to me, and then I can… I can have Byron just export it and send it to us before the meeting, so…

1092 01:47:04.080 ⇒ 01:47:13.040 Uttam Kumaran: Okay, so… Or whatever. Yeah, so let’s do… so I’ll put a slide here, which is… Data warehouse…

1093 01:47:16.730 ⇒ 01:47:28.040 Uttam Kumaran: And then… We also want to do a slide on, like, architecture… So…

1094 01:47:29.070 ⇒ 01:47:33.880 Uttam Kumaran: I think one thing we can do here is maybe I’ll ask

1095 01:47:34.330 ⇒ 01:47:45.220 Uttam Kumaran: Mustafa, if you can put together, like, a… one of the… a little bit of, like, a… or work with, marketing to put together the data architecture diagram for this client.

1096 01:47:46.550 ⇒ 01:47:50.699 Uttam Kumaran: Or, I don’t know, Sam, if that’s, like, you think more in your course, since Mustafa’s kind of new.

1097 01:47:54.090 ⇒ 01:47:55.220 Samuel Roberts: We didn’t many work days.

1098 01:47:57.230 ⇒ 01:47:57.910 Uttam Kumaran: Okay.

1099 01:47:58.560 ⇒ 01:47:59.230 Mustafa Raja: Yeah.

1100 01:48:00.460 ⇒ 01:48:03.739 Mustafa Raja: I’d actually like to shadow Sam if he’s taking a chance.

1101 01:48:03.740 ⇒ 01:48:04.270 Samuel Roberts: down.

1102 01:48:04.590 ⇒ 01:48:10.440 Uttam Kumaran: Okay, okay, great. Yeah, and then you guys just work with, work with Hannah. She can whip it together pretty quickly.

1103 01:48:11.180 ⇒ 01:48:11.750 Mustafa Raja: Okay.

1104 01:48:12.420 ⇒ 01:48:20.699 Uttam Kumaran: She’s worked with marketing to get a data platform diagram for HS.

1105 01:48:21.080 ⇒ 01:48:22.550 Uttam Kumaran: There’s no BI tool.

1106 01:48:24.710 ⇒ 01:48:33.899 Uttam Kumaran: But we have ETL, sources, warehouse… And… dbt… On the way.

1107 01:48:34.570 ⇒ 01:48:35.180 Uttam Kumaran: Cool.

1108 01:48:35.180 ⇒ 01:48:35.700 Samuel Roberts: Cool.

1109 01:48:36.080 ⇒ 01:48:40.540 Uttam Kumaran: Similarly for this, I think if you can just…

1110 01:48:41.090 ⇒ 01:48:46.340 Uttam Kumaran: If you could just do a quick bullet on, like, the sources, Damn.

1111 01:48:47.470 ⇒ 01:48:57.420 Uttam Kumaran: So, I’ll do bullets on each source, and high level of what we have, don’t have.

1112 01:48:57.600 ⇒ 01:48:58.460 Uttam Kumaran: That’s great.

1113 01:49:01.280 ⇒ 01:49:02.180 Samuel Roberts: And…

1114 01:49:02.180 ⇒ 01:49:08.480 Uttam Kumaran: And so that’s… yeah, that’s perfect. Okay, cool. Insomnia…

1115 01:49:08.660 ⇒ 01:49:12.139 Uttam Kumaran: I don’t believe Robert… I told Amber that she can just…

1116 01:49:12.630 ⇒ 01:49:17.940 Uttam Kumaran: Send an update into the channel, unless you wanted to do something differently today.

1117 01:49:18.610 ⇒ 01:49:20.580 Robert Tseng: Yeah, no, I think that’s fine for today.

1118 01:49:21.930 ⇒ 01:49:22.620 Uttam Kumaran: Okay.

1119 01:49:22.940 ⇒ 01:49:26.829 Uttam Kumaran: Eden, yeah, Ashwini started to take some modeling things.

1120 01:49:27.110 ⇒ 01:49:28.390 Uttam Kumaran: Pushing stuff out.

1121 01:49:29.110 ⇒ 01:49:33.600 Uttam Kumaran: I don’t have, like… we don’t have much of an update there on anything.

1122 01:49:34.180 ⇒ 01:49:40.220 Uttam Kumaran: What else? Urban Stems.

1123 01:49:40.680 ⇒ 01:49:49.060 Uttam Kumaran: Yeah, I guess maybe while I have you on the phone, I pitched them on… the 5K… .

1124 01:49:49.770 ⇒ 01:49:51.289 Robert Tseng: Yeah, I saw a Zach’s response.

1125 01:49:51.530 ⇒ 01:50:00.019 Uttam Kumaran: what do you think? Should I just go… go for it? I mean, I want… I kind of want him to be like, we’re down, and… but we’re gonna promise that we’ll work with you later.

1126 01:50:02.970 ⇒ 01:50:07.140 Robert Tseng: Wait, you mean you got his reply on he doesn’t want to do the retainer, right?

1127 01:50:07.140 ⇒ 01:50:14.170 Uttam Kumaran: Yeah, I know, but I… but I… I want him to, like, sign something that says, we’re gonna work with you after 3 months, like, in a bigger way.

1128 01:50:14.870 ⇒ 01:50:15.810 Robert Tseng: Oh, I see.

1129 01:50:17.790 ⇒ 01:50:23.010 Uttam Kumaran: That would be my only, like, pushback, is I’m like, Well…

1130 01:50:23.010 ⇒ 01:50:24.410 Robert Tseng: That’s… that’s tough. It’s…

1131 01:50:25.680 ⇒ 01:50:26.130 Uttam Kumaran: Well, like…

1132 01:50:26.130 ⇒ 01:50:29.290 Robert Tseng: I feel like we should just switch the lights off on them for a month.

1133 01:50:32.560 ⇒ 01:50:36.710 Uttam Kumaran: Let me see, yeah, this is where, like, yeah, you’re more ruthless than I am.

1134 01:50:38.080 ⇒ 01:50:40.640 Uttam Kumaran: I mean, yeah, I, I, I mean, I agree, like…

1135 01:50:40.910 ⇒ 01:50:50.760 Robert Tseng: Because he’s gonna sign that, and their budget is gonna not… his budget’s not out for 2026. Like, I feel like it’s hard to hold him to… to that agreement, because he’s just gonna be like, well, the budget changed.

1136 01:50:56.590 ⇒ 01:50:57.250 Uttam Kumaran: Yeah.

1137 01:50:59.310 ⇒ 01:51:05.919 Robert Tseng: Yeah, like, I kind of want him to know that, like, this is the bare minimum he needs to be able to go to bat for us on, and then, like…

1138 01:51:06.170 ⇒ 01:51:06.930 Robert Tseng: you know.

1139 01:51:07.930 ⇒ 01:51:14.630 Uttam Kumaran: Well, I think what… you know what he’s trying to do, is he’s like… I think he’s… that budget or whatever includes Emily and all these people.

1140 01:51:15.220 ⇒ 01:51:15.590 Robert Tseng: Yes.

1141 01:51:15.590 ⇒ 01:51:17.910 Uttam Kumaran: So he’s thinking about it holistically, and I’m like…

1142 01:51:18.440 ⇒ 01:51:19.979 Uttam Kumaran: I get that he wants problems.

1143 01:51:19.980 ⇒ 01:51:21.369 Robert Tseng: Hour budget, yeah.

1144 01:51:21.370 ⇒ 01:51:25.840 Uttam Kumaran: he wants to remove some of them and free up the budget, but, like, I kind of… I don’t know, I would…

1145 01:51:26.000 ⇒ 01:51:27.330 Uttam Kumaran: I’m trying to, like…

1146 01:51:27.600 ⇒ 01:51:32.250 Uttam Kumaran: trying to get him to say, like, can you sign some… I don’t know. Or we can say, like, look.

1147 01:51:32.440 ⇒ 01:51:39.770 Uttam Kumaran: we need some minimums, because for us, like, it’s… we have to get organized around it, too, you know? And plan, and so…

1148 01:51:39.770 ⇒ 01:51:41.690 Robert Tseng: Yeah, because we don’t do staff aug.

1149 01:51:41.850 ⇒ 01:51:51.079 Robert Tseng: That’s basically what he’s gonna ask for, which is like, well, staff aug with no limits, and you want a single all-in price? Well, like, no, we can’t guarantee that.

1150 01:51:51.600 ⇒ 01:51:52.210 Uttam Kumaran: Yeah.

1151 01:51:56.780 ⇒ 01:51:58.790 Robert Tseng: Yeah, I mean, like, I think that it’s, like.

1152 01:51:58.930 ⇒ 01:52:07.620 Robert Tseng: we’re either gonna rotate our staff onto… I mean, I don’t know, do you feel like we need to stretch to him for the cash, or I feel like we don’t, like, we can…

1153 01:52:08.310 ⇒ 01:52:12.989 Uttam Kumaran: No, I don’t want to stretch them for the cash, but I want to work with them next year.

1154 01:52:13.190 ⇒ 01:52:14.730 Robert Tseng: Yeah. You know, so…

1155 01:52:17.060 ⇒ 01:52:22.609 Uttam Kumaran: I mean, I can ask him for a lower minimum, but then I want to let him… I can ask him for a lower minimum.

1156 01:52:29.230 ⇒ 01:52:36.989 Uttam Kumaran: For me, the minimum is just, like, a signifier that we’re, like, why don’t we just charge him $200 an hour, and then it’s kind of like, as the needs come.

1157 01:52:37.110 ⇒ 01:52:37.740 Robert Tseng: You can just go.

1158 01:52:37.740 ⇒ 01:52:39.879 Uttam Kumaran: No, I mean, that’s what basically he said.

1159 01:52:40.570 ⇒ 01:52:41.120 Robert Tseng: Yeah.

1160 01:52:41.260 ⇒ 01:52:47.429 Robert Tseng: Well, I mean, that’s… if you’re looking for an alternative, I don’t… I don’t know what a minimum under $5K would look like, you know?

1161 01:52:53.480 ⇒ 01:52:58.840 Uttam Kumaran: Okay, I mean, then the… then basically I’ll say… Sure.

1162 01:53:00.340 ⇒ 01:53:00.930 Robert Tseng: But…

1163 01:53:01.760 ⇒ 01:53:03.950 Uttam Kumaran: Well, I don’t, but I don’t know, I mean, it’s like…

1164 01:53:04.470 ⇒ 01:53:09.610 Uttam Kumaran: I mean, I would like to keep charging them for work we do. There’s no way…

1165 01:53:10.140 ⇒ 01:53:13.570 Uttam Kumaran: They figure it out without us, but also…

1166 01:53:13.570 ⇒ 01:53:14.150 Robert Tseng: Yeah.

1167 01:53:17.980 ⇒ 01:53:21.659 Uttam Kumaran: Okay. I mean, like, I don’t… the reason I don’t want to cut them off…

1168 01:53:21.860 ⇒ 01:53:23.880 Uttam Kumaran: is, like, I just wanna, like.

1169 01:53:24.160 ⇒ 01:53:26.399 Uttam Kumaran: Get the next scope, you know?

1170 01:53:30.020 ⇒ 01:53:30.610 Robert Tseng: Yeah.

1171 01:53:33.460 ⇒ 01:53:40.810 Robert Tseng: Well, then how about we… how about we do it, like, okay, we’ll do hourly with you, but we want you to work with us to build out that next scope?

1172 01:53:42.480 ⇒ 01:53:48.749 Robert Tseng: Like, maybe it’s like, we do the hourly, we don’t charge him for his time to kind of loop us in on the next scope.

1173 01:53:49.000 ⇒ 01:53:57.289 Robert Tseng: Okay. But, like, he needs… yeah, so something like that, where, like, we’re basically holding to him to, like, he basically has to, like, help us get our next contract.

1174 01:53:57.870 ⇒ 01:54:02.770 Uttam Kumaran: Okay, so… I will send… okay, we’re good with the no minimums.

1175 01:54:03.390 ⇒ 01:54:12.450 Uttam Kumaran: But, as a concession, And for no additional charge for our hours to do so, we’d like to…

1176 01:54:13.950 ⇒ 01:54:19.809 Uttam Kumaran: We’d like to schedule time to work with you on the… on the scope.

1177 01:54:20.350 ⇒ 01:54:22.550 Uttam Kumaran: On the next phase scope. Okay.

1178 01:54:22.550 ⇒ 01:54:27.029 Robert Tseng: Yeah, because, like, we just tell them, like, this hourly, like, bespoke hourly.

1179 01:54:27.720 ⇒ 01:54:28.170 Uttam Kumaran: That’s what we do.

1180 01:54:28.770 ⇒ 01:54:35.790 Robert Tseng: It’s not something we do. We can do it to the end of the year, just to kind of help you smooth it over, but, like, the goal is to kind of, like.

1181 01:54:36.170 ⇒ 01:54:39.609 Robert Tseng: You know, obviously, go for… go for the renewal. So, like.

1182 01:54:40.070 ⇒ 01:54:49.860 Robert Tseng: We’re willing to do the hourly through the end of the year, if you’re able to, like, kind of co-author this next scope with us, or, you know, however you want to, like.

1183 01:54:50.270 ⇒ 01:54:52.550 Robert Tseng: We can talk about the wording afterwards.

1184 01:54:53.100 ⇒ 01:54:53.700 Uttam Kumaran: Okay.

1185 01:54:54.700 ⇒ 01:54:55.250 Robert Tseng: Yeah.

1186 01:54:56.950 ⇒ 01:55:01.249 Robert Tseng: And we’re kind of like, it’s not like… yeah, I think there’s a good way to phrase it, where it’s like…

1187 01:55:01.250 ⇒ 01:55:01.740 Uttam Kumaran: Yeah. Basically.

1188 01:55:02.820 ⇒ 01:55:05.009 Robert Tseng: Free discovery, like, for the next…

1189 01:55:05.200 ⇒ 01:55:10.020 Robert Tseng: For the next phase. But, like, we just want you to know that that’s our intention in signing this.

1190 01:55:11.150 ⇒ 01:55:11.680 Uttam Kumaran: Okay.

1191 01:55:12.260 ⇒ 01:55:13.110 Robert Tseng: Okay.

1192 01:55:13.110 ⇒ 01:55:13.820 Uttam Kumaran: Perfect.

1193 01:55:14.260 ⇒ 01:55:18.190 Uttam Kumaran: Cool,

1194 01:55:21.500 ⇒ 01:55:28.050 Uttam Kumaran: What else? Readme, yeah, I handed over…

1195 01:55:28.990 ⇒ 01:55:31.700 Uttam Kumaran: what Alicia asked for to Mustafa, so…

1196 01:55:31.700 ⇒ 01:55:32.280 Robert Tseng: Great.

1197 01:55:32.720 ⇒ 01:55:36.820 Uttam Kumaran: all, like, I think… thanks, Mustafa, you can… You could crack on.

1198 01:55:37.120 ⇒ 01:55:43.390 Uttam Kumaran: I think as soon as you’re… maybe after another week or so, I’ll… I can loop Mustafa into this channel.

1199 01:55:44.360 ⇒ 01:55:44.920 Robert Tseng: Sweet.

1200 01:55:47.320 ⇒ 01:56:01.439 Uttam Kumaran: And then I also mentioned to Amber to start looping in Casey on insomnia-related analysis, and then to start looping in Casey on… I mean, Mustafa on Honey Stinger-related analysis, so we have some redundancy there.

1201 01:56:01.710 ⇒ 01:56:04.109 Uttam Kumaran: So I feel pretty good about those.

1202 01:56:05.080 ⇒ 01:56:08.550 Uttam Kumaran: all the other clients, like, I don’t know, feel pretty good on.

1203 01:56:12.540 ⇒ 01:56:15.030 Uttam Kumaran: Anything else we wanna chat about?

1204 01:56:15.610 ⇒ 01:56:17.190 Uttam Kumaran: It’s kind of mainly what I had.

1205 01:56:18.260 ⇒ 01:56:23.030 Robert Tseng: Yeah, I mean, to me, like, I’m gonna wake up in a few hours, I’m gonna take the Honey Stinger call.

1206 01:56:23.300 ⇒ 01:56:30.209 Robert Tseng: I’m gonna catch up with CERF on Remo. We’re just trying to put together that project doc, and I want to send that proposal out.

1207 01:56:30.460 ⇒ 01:56:33.450 Robert Tseng: If not end of day today, probably over the weekend.

1208 01:56:33.560 ⇒ 01:56:36.719 Uttam Kumaran: Can you send that to me? Or whenever? Or put it, or put it in the channel?

1209 01:56:36.720 ⇒ 01:56:43.649 Robert Tseng: work on it. I basically just needed… I need him to tell me that, like, we’re gonna have to build, like, a Gantt chart kind of thing, just so…

1210 01:56:43.650 ⇒ 01:56:45.050 Uttam Kumaran: Tell him to build it, dude.

1211 01:56:47.380 ⇒ 01:56:49.290 Uttam Kumaran: Tell him to take the first crack at it.

1212 01:56:49.540 ⇒ 01:56:54.869 Robert Tseng: Did, Awash, did… did Surf reach out to you already? Because I did ask him to work with you.

1213 01:56:54.870 ⇒ 01:57:01.380 Awaish Kumar: Yeah, regarding creating some data flow diagrams…

1214 01:57:01.970 ⇒ 01:57:02.680 Robert Tseng: Yeah.

1215 01:57:03.780 ⇒ 01:57:04.790 Awaish Kumar: Do you…

1216 01:57:05.540 ⇒ 01:57:17.559 Awaish Kumar: Yeah, he did ask me, but I asked him about his, like, availability, but I didn’t get the reply. I will ask him again on, like, because he’s not…

1217 01:57:17.560 ⇒ 01:57:20.890 Robert Tseng: I’m drafting a message, I’ll nudge him right now. Okay, yeah.

1218 01:57:21.070 ⇒ 01:57:24.180 Robert Tseng: So that’s that, and then,

1219 01:57:25.100 ⇒ 01:57:38.219 Robert Tseng: Yeah, actually, with Insomnia, I’m trying to… trying to steer them towards… because they do have tracking and tagging needs, so I’m trying to get Zoran in. So, that’s another thing I’m actively talking to Amrita about.

1220 01:57:40.760 ⇒ 01:57:42.110 Uttam Kumaran: Okay, okay, great.

1221 01:57:42.110 ⇒ 01:57:42.750 Robert Tseng: Yeah.

1222 01:57:42.910 ⇒ 01:57:51.099 Robert Tseng: Just because, like, the pace of the insomnia analysis is not our fault, it’s just, like, you know, they’re so low resource, they can’t action the stuff.

1223 01:57:51.170 ⇒ 01:58:05.389 Robert Tseng: Matt obviously has slow turnarounds for you guys, and then also, like, Birdie is gonna… is already queuing up, like, the recommendations from Amber, but it’s not gonna be deployed for another week and a half or something. So, it’s just like…

1224 01:58:05.520 ⇒ 01:58:10.660 Robert Tseng: I don’t know, I just… I’m just trying to have more… have more irons in the fire here.

1225 01:58:10.970 ⇒ 01:58:11.560 Uttam Kumaran: Okay.

1226 01:58:12.310 ⇒ 01:58:12.860 Robert Tseng: Yeah.

1227 01:58:14.240 ⇒ 01:58:19.040 Uttam Kumaran: I mean, we also have made no inroads onto the, like, data platform.

1228 01:58:19.660 ⇒ 01:58:27.159 Uttam Kumaran: Stuff for them, so… That’s also tough, because Casey’s still, like, working on things daily.

1229 01:58:28.430 ⇒ 01:58:33.909 Robert Tseng: Yeah, I didn’t end up making that spreadsheet for him. That’s not gonna happen this week.

1230 01:58:33.910 ⇒ 01:58:35.250 Uttam Kumaran: That’s fine. Yeah, yeah, yeah.

1231 01:58:37.190 ⇒ 01:58:38.680 Uttam Kumaran: Okay, cool.

1232 01:58:38.980 ⇒ 01:58:42.520 Uttam Kumaran: Yeah, I feel good otherwise.

1233 01:58:46.830 ⇒ 01:58:47.450 Robert Tseng: Okay.

1234 01:58:49.090 ⇒ 01:58:53.009 Uttam Kumaran: Okay, great. Alright, I’ll talk to everybody else on… on Slack.

1235 01:58:54.140 ⇒ 01:58:54.810 Robert Tseng: Thanks.

1236 01:58:54.900 ⇒ 01:58:55.769 Mustafa Raja: Thank you. Okay.

1237 01:58:55.770 ⇒ 01:58:56.470 Uttam Kumaran: Thank you.

1238 01:58:56.600 ⇒ 01:58:57.370 Uttam Kumaran: Talk to you soon.

1239 01:58:57.370 ⇒ 01:58:58.040 Samuel Roberts: How many?