Meeting Title: Uttam Kumaran Date: 2025-03-18 Meeting participants: Luke Daque, Uttam Kumaran, Awaish Kumar, Caio Velasco


WEBVTT

1 00:03:24.820 00:03:25.430 Caio Velasco: Hey?

2 00:03:25.620 00:03:26.490 Caio Velasco: I’m good.

3 00:03:30.660 00:03:32.999 Caio Velasco: for example, just a just a quick thing

4 00:03:33.570 00:03:35.610 Uttam Kumaran: Yeah. What? What say that? Again?

5 00:03:35.850 00:03:43.260 Caio Velasco: I was just gonna say that now I just saw that Robert was tagging me on on some like cogs related work. And I was like. But it’s not me.

6 00:03:46.020 00:04:01.780 Uttam Kumaran: You’re gonna get hit with a hundred things I’m telling you. This is what’s gonna this is what makes a good engineer to a great engineer. You need to have the triage process in your brain. You can say, Yeah, we’ll take care of it. Y’all, that’s all they need to hear is, yeah, we’ll take care of it.

7 00:04:02.390 00:04:04.400 Uttam Kumaran: They don’t need to hear the solution.

8 00:04:05.020 00:04:17.070 Uttam Kumaran: That’s what I’m saying, because this is what I this is like, totally what happens, and I feel like I’m so I know I just. I know I’m comfortable with this, because I’ve I’ve just. This is my life for like the last 6 years.

9 00:04:17.170 00:04:23.869 Uttam Kumaran: But the client they ultimately don’t care how you solve. They just wanna know that you’re there.

10 00:04:24.420 00:04:47.560 Uttam Kumaran: It’s like they wanna know they want the warm hug of. I got it. Then the timer starts, though, right? But that’s it. Like the problems are gonna stream and the the whole project management function. Is there to be a shield between us? And I say us because I I’m more of engineer like us and the client, the project manager in the middle

11 00:04:47.710 00:04:55.779 Uttam Kumaran: stuff is still gonna find its way around to you like client may tag you right ideally. The project manager should be the one to triage everything.

12 00:04:56.120 00:05:01.229 Uttam Kumaran: But look, if the triage process takes 3 days and it’s a 1 line thing.

13 00:05:01.900 00:05:05.750 Uttam Kumaran: The reason we are we’re in business is because we get it done in one day.

14 00:05:06.160 00:05:23.987 Uttam Kumaran: right? And so there we will have the opportunity to find. What stuff can we do? Really, quickly, versus what stuff is coming later, but for anything that’s flowing in. Just give the thumbs up or say, cool. Let’s we’ll get a triage. We’ll get a triage, or you can quickly just create an issue, and from

15 00:05:24.750 00:05:30.410 Uttam Kumaran: from slack like. If you if you click in slack, I’ll show you if you just click like

16 00:05:31.340 00:05:34.619 Uttam Kumaran: if you go to Javi, and then you click like,

17 00:05:37.830 00:05:39.313 Uttam Kumaran: I don’t know. Let’s say,

18 00:05:40.590 00:05:44.240 Uttam Kumaran: one of these here, you could just say, create an issue in linear.

19 00:05:44.900 00:05:47.690 Uttam Kumaran: And you could just say I created issue. Good! It’s there. It’s tracked

20 00:05:48.360 00:05:50.657 Uttam Kumaran: that will buy you some time

21 00:05:51.040 00:05:52.530 Caio Velasco: I get it. Okay.

22 00:05:53.000 00:06:00.630 Uttam Kumaran: That’s just the way it’s gonna go for a bit. It’s just gonna go like that. I think the the thing I’m trying to get to. This is what happens naturally. And data is that

23 00:06:00.950 00:06:06.830 Uttam Kumaran: it’s like A, we’re customer service. So customer is gonna have a hundred problems.

24 00:06:07.620 00:06:14.870 Uttam Kumaran: It’s our job to just assure them that we’re gonna fix it. And then we, as a team, work on like what to do, and we go one by one by one.

25 00:06:15.150 00:06:16.210 Uttam Kumaran: you know, but

26 00:06:16.310 00:06:43.569 Uttam Kumaran: it’s tough right now. We’re in this like growing moment. So for me, ultimately, like, I want to make sure that one. You feel you have enough time. So certainly I mean, like I think we have to do stand ups like the Pm’s. Definitely need that. If you need more time, take it like in the short term period. I think, as you need it, just let me know how much. Just let me know how much more time you don’t need to go through Steven, to get that approved or anything. Just take as much time as you need to get up to speed.

27 00:06:44.000 00:06:51.860 Uttam Kumaran: I think over time I ideally want. I want all of our engineers to be able to take on 2 or 3 clients with the 40 h that they have.

28 00:06:52.050 00:06:58.879 Uttam Kumaran: So keep that in mind, like I’m not. I don’t want to be like you only have 10 h here and 20. It doesn’t like engineering doesn’t work like that.

29 00:07:00.430 00:07:01.399 Caio Velasco: Mine was a bit like this.

30 00:07:01.400 00:07:03.959 Uttam Kumaran: No, I know I’m in a bind because

31 00:07:04.400 00:07:13.070 Uttam Kumaran: the business is measured like that right? And we build clients like that. So it’s a real hard thing for me. Take as much time as you need.

32 00:07:13.280 00:07:26.020 Uttam Kumaran: I want you to be really the lead on Javi at minimum. And then, ideally, I want to be able to say cool with 20 h. You feel comfortable with Javi. Let me like over time.

33 00:07:26.200 00:07:34.199 Uttam Kumaran: Let’s see how we can get you. One more client. And ideally, we’re each on like at least 2 clients, maybe 2 primary, 2 secondary.

34 00:07:34.340 00:07:36.410 Uttam Kumaran: That builds some sort of layering

35 00:07:36.790 00:07:53.169 Uttam Kumaran: that’s like my, that’s like my vision here. Right? So, even with a wish, a wish is really just on Eden. But then I moved him also to support on Javi between the 5 of us, or whatever we can cover everything. Another 2 clients is sort of how I’m thinking right?

36 00:07:53.560 00:07:55.549 Uttam Kumaran: So take as much time as you want.

37 00:07:55.740 00:07:59.520 Uttam Kumaran: and just get up to speed as fast as you can.

38 00:07:59.840 00:08:06.810 Uttam Kumaran: If we go. If you go over them on time. It’s fine. It’s fine in the short term, and we’ll we can. We can take. We can take the head.

39 00:08:08.340 00:08:11.920 Uttam Kumaran: But that’s that’s probably what I would say. There, so that’s fine.

40 00:08:12.710 00:08:23.080 Caio Velasco: Okay, good. I think if you’re more comfortable with this, like focusing on Javier, knowing that tomorrow morning I can spend the whole day and not just like 30 min here, 30 min there

41 00:08:23.080 00:08:31.590 Uttam Kumaran: Sorry about that shit. I’m sorry if I didn’t make that clear before like, and I didn’t. I didn’t make that clear before, so that is on that is on me completely.

42 00:08:31.890 00:08:38.819 Uttam Kumaran: But it’s also like it’s tough because it’s just tough, because the the business is run on hourly allocation, right?

43 00:08:38.940 00:08:45.189 Uttam Kumaran: Partly the business, like I run the business like, it’s data. So I sort of know how many hours everyone has.

44 00:08:45.330 00:08:50.380 Uttam Kumaran: But I also know that, like I can’t. It’s impossible to be like 1 h here, 2 h there, 3 h there.

45 00:08:51.250 00:08:57.260 Uttam Kumaran: But I also know that roughly, for us to break even on Javi. It’s about 20 to 30 h total.

46 00:08:57.500 00:09:00.560 Uttam Kumaran: So if I tell you, hey, go spend 40 h.

47 00:09:01.300 00:09:16.570 Uttam Kumaran: We’re not. We’re not gonna make any money. So then I have like 2 things on my shoulder. I’m like, we just don’t make money here. But then also I want you to get up to speed so. But there will be some taxes you ramp up, and I’m okay with that. So feel free to spend as much time as you need. Ideally, I would love

48 00:09:16.770 00:09:22.700 Uttam Kumaran: for you to at least be able to take on in the short term. And I’m telling this to everybody is at least 2 clients. So consider that

49 00:09:23.850 00:09:27.199 Uttam Kumaran: I think you’re on Javi. I think pool parts is also

50 00:09:27.510 00:09:33.329 Uttam Kumaran: some work may come down the line. If no work comes down the line on pool parts. I have other work that we need to do.

51 00:09:33.570 00:09:35.349 Uttam Kumaran: and so I think

52 00:09:35.650 00:09:40.420 Uttam Kumaran: you can. I think that’s a that’s a good sort of way to to balance everything

53 00:09:41.880 00:09:48.739 Caio Velasco: Okay, okay, sounds good. Sounds good. Okay, perfect. Perfect.

54 00:09:49.265 00:09:55.860 Caio Velasco: How can we check this channel problems? A phone situation

55 00:09:55.860 00:10:04.689 Uttam Kumaran: Yeah, yeah. So you want to share, you want to share your screen, or or I think, like one. Maybe I I think what would be good 1st is to just tell them on.

56 00:10:04.860 00:10:08.059 Uttam Kumaran: hey, this change is filtering out 60,000 rows.

57 00:10:08.950 00:10:14.189 Uttam Kumaran: I think what we’re gonna do instead is just move these out of the other category.

58 00:10:14.930 00:10:20.110 Uttam Kumaran: And the other thing we’re gonna do is these tickets stop in July.

59 00:10:20.640 00:10:24.380 Uttam Kumaran: so don’t worry about it as long as you’re not filtering to July. You’re not going to see them.

60 00:10:25.200 00:10:31.720 Uttam Kumaran: If you can send that to him on right now, I think you’ll get the thumbs up, and that’ll be the Pr. That’ll be the change.

61 00:10:33.400 00:10:34.349 Uttam Kumaran: Do you see that?

62 00:10:34.350 00:10:35.030 Uttam Kumaran: Yep.

63 00:10:36.163 00:10:43.440 Caio Velasco: But then but let me see if I got it right. So the on pro the July thing is the

64 00:10:43.620 00:10:48.160 Caio Velasco: is the oldest ticket, not the earliest

65 00:10:50.860 00:10:54.899 Uttam Kumaran: Oh, so there are tickets more recent than that

66 00:10:55.530 00:10:57.820 Caio Velasco: Yeah, yeah, yeah, for sure. For sure.

67 00:10:57.960 00:11:01.139 Uttam Kumaran: Oh, then we should tell him, because he said the phone. Well.

68 00:11:02.930 00:11:04.330 Uttam Kumaran: then, you should tell him that

69 00:11:05.710 00:11:06.150 Caio Velasco: Okay.

70 00:11:06.150 00:11:10.629 Uttam Kumaran: There’s both phone. And here the here are the most recent dates for phone and SMS.

71 00:11:12.680 00:11:18.650 Uttam Kumaran: If we filter these out, we’re gonna filter these and 60,000 more. What do you want to do? You just send that.

72 00:11:19.980 00:11:22.120 Uttam Kumaran: We’ll be good. We’ll it’ll buy some time

73 00:11:29.480 00:11:30.320 Caio Velasco: Do

74 00:11:40.810 00:11:46.960 Caio Velasco: let me check so that we can just see if it’s what I’m saying is correct.

75 00:12:40.930 00:12:46.620 Caio Velasco: So maybe in broad, the

76 00:13:00.850 00:13:04.680 Caio Velasco: okay. So for phone. The most recent one is from today

77 00:13:04.850 00:13:09.360 Caio Velasco: and from SMS. The most recent one is actually from

78 00:13:10.450 00:13:14.230 Caio Velasco: is this month, right? So last year may last year, I think

79 00:13:14.640 00:13:18.324 Uttam Kumaran: Can you? Can you just send that? Just send just what you said to me.

80 00:13:18.760 00:13:19.669 Uttam Kumaran: Come on.

81 00:13:20.820 00:13:21.590 Caio Velasco: Okay.

82 00:13:24.330 00:13:28.359 Uttam Kumaran: And then, while while we’re on this, I think maybe it’s helpful

83 00:13:30.220 00:13:32.310 Uttam Kumaran: like, what did they mean by other

84 00:13:37.070 00:13:40.299 Uttam Kumaran: like when he said, the channel is going to other.

85 00:13:42.650 00:13:44.529 Uttam Kumaran: Is that something in Meta Base?

86 00:13:45.150 00:13:47.469 Uttam Kumaran: So I don’t see a category field here.

87 00:13:48.420 00:13:52.079 Uttam Kumaran: Okay, I’ll let you. I’ll let you finish sending. Send us some on first, st we can talk about the next thing. Yeah.

88 00:13:52.080 00:13:55.510 Caio Velasco: Okay, check anywhere. Okay, fine.

89 00:15:55.920 00:16:03.740 Caio Velasco: So as I understand correctly that this, if I think it is create, if I have a a role for a phone

90 00:16:04.410 00:16:09.110 Caio Velasco: value in the Channel call is that a ticket was created by Via phone

91 00:16:09.540 00:16:16.200 Uttam Kumaran: I think so. Maybe what’s also helpful is, pull the pull the id of the of those latest ones and send it to him.

92 00:16:16.720 00:16:17.340 Caio Velasco: Okay.

93 00:16:18.290 00:16:25.350 Uttam Kumaran: Yeah, but basically, yes, which which contradicts what he said. So it’s like.

94 00:16:30.970 00:16:46.190 Uttam Kumaran: if I can just do this all day, I would love to do this. I’ll just come. Do engineering work. We can work together. This is what I love, but I couldn’t do it before, because I’m running all the stand ups right? So I would do. 3 h of stand ups, do another 2 h of client meetings.

95 00:16:46.520 00:16:53.350 Uttam Kumaran: And then I’m like, I just wanna I just wanna do dbt work. That’s all I wanna do.

96 00:16:55.810 00:17:01.810 Uttam Kumaran: That’s why I started that. You know, it’s funny is all I want to do is what I started doing is just dbt, work

97 00:17:04.880 00:17:07.069 Caio Velasco: What is the ticket type?

98 00:17:12.140 00:17:12.819 Caio Velasco: Everyone?

99 00:17:25.230 00:17:27.460 Caio Velasco: you can’t guarantee.

100 00:18:20.790 00:18:26.079 Caio Velasco: Okay? Yeah. Because I’m then I’m not playing with the with the query to get to think. But then

101 00:18:26.490 00:18:29.379 Caio Velasco: I I get possibly don’t have to focus

102 00:18:30.580 00:18:31.830 Uttam Kumaran: No, you’re good. You’re good.

103 00:18:58.530 00:18:59.830 Caio Velasco: See if it matches.

104 00:21:02.810 00:21:03.520 Caio Velasco: I don’t know.

105 00:21:11.250 00:21:13.790 Caio Velasco: Sandwich.

106 00:21:43.740 00:21:44.979 Caio Velasco: Okay? Done.

107 00:21:52.290 00:21:55.849 Caio Velasco: But it just basically said that tickets with phone, you still receive it.

108 00:21:55.990 00:22:03.940 Caio Velasco: And for SMS. It stopped within 2024, and then I put the both features

109 00:22:04.370 00:22:09.490 Caio Velasco: screenshots with the ticket Id for both SMS and the other one

110 00:22:09.880 00:22:10.580 Uttam Kumaran: Okay.

111 00:22:20.630 00:22:32.829 Caio Velasco: Yeah, because it’s why are we? Yeah, that’s a good question. Why are we still receiving phones? It was just convenient. That’s why I think, he said. It’s missed calls, maybe because you can call it, and

112 00:22:33.820 00:22:36.339 Uttam Kumaran: Oh, maybe people are calling, and nobody picks up

113 00:22:37.220 00:22:37.940 Caio Velasco: Okay.

114 00:22:38.960 00:22:44.959 Uttam Kumaran: Yeah. But then again, I want him to sort of tell us what the deal is.

115 00:22:46.920 00:22:47.580 Uttam Kumaran: You know

116 00:22:47.580 00:22:47.949 Caio Velasco: Okay.

117 00:22:53.890 00:22:57.280 Caio Velasco: okay, what would be an extended

118 00:22:58.160 00:23:03.729 Uttam Kumaran: So then for okay, let’s just take a look at like what’s going on in

119 00:23:06.040 00:23:08.079 Uttam Kumaran: like, what’s going on here in

120 00:23:08.420 00:23:10.729 Uttam Kumaran: what is this is this in gorgeous? Yeah.

121 00:23:12.870 00:23:17.620 Uttam Kumaran: So there’s this other. See? This is probably what they got hooked on was like, what is this?

122 00:23:18.460 00:23:21.230 Uttam Kumaran: But if I go check out what this is.

123 00:23:25.690 00:23:29.360 Uttam Kumaran: Oh, okay, just by channel.

124 00:23:32.730 00:23:36.740 Uttam Kumaran: So if we go to

125 00:23:37.270 00:23:41.079 Caio Velasco: In the beginning, also didn’t understand why we had like borders.

126 00:23:41.230 00:23:45.739 Caio Velasco: the fact tickets. They have all the values in the Channel quality

127 00:23:45.740 00:23:50.920 Uttam Kumaran: Yeah. So that’s what I’m sort of like, what I don’t know what’s going on, because

128 00:23:51.180 00:23:53.290 Uttam Kumaran: it’s td, dot channel here.

129 00:24:01.350 00:24:03.189 Uttam Kumaran: Yeah, right? It’s just channel

130 00:24:03.820 00:24:06.920 Caio Velasco: It’s just channel. Yeah, there’s no I

131 00:24:07.070 00:24:10.180 Caio Velasco: grouping by, or something, you know.

132 00:24:11.430 00:24:14.640 Uttam Kumaran: Okay, so let’s go back

133 00:24:25.855 00:24:35.049 Caio Velasco: Yeah. And then also remember that it’s I don’t know if, since you said that, you’re cloning, I don’t think the problem would be in the fact. It is in analytics, right? Because it’s the same

134 00:24:35.940 00:24:41.569 Uttam Kumaran: Yeah, you’re right. I mean, I don’t. I don’t think that should be a problem. But we can go check this also.

135 00:24:43.820 00:24:50.459 Caio Velasco: Interesting. Maybe database automatically put everything that’s super small in in order

136 00:25:54.300 00:25:54.990 Uttam Kumaran: What

137 00:26:00.350 00:26:02.079 Uttam Kumaran: I don’t even see other in here

138 00:26:02.430 00:26:03.940 Caio Velasco: Yeah, exactly interesting.

139 00:26:10.580 00:26:15.249 Caio Velasco: And the question, the the sequel on the right was created by us, or is automatic

140 00:26:15.250 00:26:20.279 Uttam Kumaran: It’s created by metabase. This is like what metabase hits the warehouse with

141 00:26:20.690 00:26:21.280 Caio Velasco: Okay.

142 00:26:22.470 00:26:24.980 Uttam Kumaran: And then, okay.

143 00:26:30.820 00:26:33.349 Uttam Kumaran: email, contact form.

144 00:26:36.780 00:26:39.600 Uttam Kumaran: oh, email, contact, form, other

145 00:26:50.970 00:26:51.520 Caio Velasco: Cool.

146 00:26:52.440 00:26:59.109 Caio Velasco: Yeah, for some reason they are getting like the at least, probably I don’t know have an idea, but it seems that they have like a threshold

147 00:26:59.490 00:27:00.270 Caio Velasco: that

148 00:27:00.270 00:27:01.540 Uttam Kumaran: Yeah, I don’t know.

149 00:27:03.700 00:27:07.490 Uttam Kumaran: Well, let’s see. Like, if I put the where clause, what happens here?

150 00:27:14.690 00:27:16.780 Uttam Kumaran: Oh, okay.

151 00:27:25.250 00:27:31.079 Uttam Kumaran: Oh, it’s just putting it into other

152 00:27:31.760 00:27:37.610 Caio Velasco: Yeah, that’s what I did this. That’s what I’m trying to say that maybe there’s an automatic thing

153 00:27:59.580 00:28:00.500 Uttam Kumaran: Oh!

154 00:28:09.340 00:28:10.490 Uttam Kumaran: So!

155 00:28:12.260 00:28:13.470 Luke Daque: Hey, guys.

156 00:28:14.190 00:28:14.890 Uttam Kumaran: Hey!

157 00:28:14.890 00:28:15.690 Caio Velasco: Hey! Man!

158 00:28:16.220 00:28:19.009 Uttam Kumaran: Okay, I think this is a good solution. I’m gonna say, this is done.

159 00:28:21.540 00:28:22.230 Uttam Kumaran: I think

160 00:28:23.120 00:28:24.009 Caio Velasco: Why.

161 00:28:24.230 00:28:33.370 Caio Velasco: no, no, it’s perfect. But why did we have the the workload is because they’re just filtering to check the the how the dashboard works

162 00:28:34.800 00:28:41.359 Uttam Kumaran: In. No, sometimes people like, if it’s too small of a percentage.

163 00:28:41.490 00:28:44.660 Uttam Kumaran: it’s not worth keeping in there. But this is the thing where it’s like.

164 00:28:45.240 00:28:48.410 Uttam Kumaran: fundamentally, this is not our problem.

165 00:28:48.990 00:28:51.809 Uttam Kumaran: right? This is a data analyst problem.

166 00:28:52.080 00:28:55.079 Uttam Kumaran: But Robert is probably presenting this in a meeting

167 00:28:55.200 00:28:57.450 Uttam Kumaran: where they said, Figure out what’s other?

168 00:28:58.210 00:29:04.510 Uttam Kumaran: And then we now spend all this time when it’s like yo, just remove the thing here and ship it like, get it out.

169 00:29:04.780 00:29:12.979 Uttam Kumaran: This is not on us like, how would we know that these aren’t supposed to be there? And, in fact, there is still phone coming in right

170 00:29:13.180 00:29:13.960 Caio Velasco: Yeah.

171 00:29:14.470 00:29:20.920 Uttam Kumaran: So I’m gonna hit done. And I’m gonna hit save. And I’m gonna say that

172 00:29:21.850 00:29:25.149 Uttam Kumaran: I’m gonna go back here. I’m gonna say.

173 00:29:25.400 00:29:29.250 Uttam Kumaran: this looks good to me. Like

174 00:29:30.410 00:29:33.299 Uttam Kumaran: I’m gonna I’m basically gonna reply in

175 00:29:33.300 00:29:40.970 Caio Velasco: Do you think you, do you think you should also remove the the where clause 1st to to even show that there is more not necessarily

176 00:29:41.820 00:29:43.380 Uttam Kumaran: Oh, meaning?

177 00:29:44.240 00:29:47.250 Uttam Kumaran: Well, right now, the where clause is just on the ticket creation date.

178 00:29:49.200 00:29:49.950 Uttam Kumaran: Yeah.

179 00:29:50.070 00:29:50.660 Uttam Kumaran: So I’m

180 00:29:50.660 00:29:53.000 Caio Velasco: They’re looking at the previous 15 days, I guess.

181 00:29:53.000 00:30:03.689 Uttam Kumaran: Yeah, I’m gonna say, like, this is, this is what this looks like for the last 15 days. Do you want to send your message, or here I’m gonna send it. You send this one to send this. I’ll I’ll send you this screenshot

182 00:30:03.690 00:30:04.789 Caio Velasco: Okay, I can

183 00:30:05.170 00:30:06.310 Uttam Kumaran: Alongside your message

184 00:30:10.560 00:30:13.740 Caio Velasco: Let me. I have the old screenshot

185 00:30:15.240 00:30:16.960 Uttam Kumaran: Yeah, send this and say.

186 00:30:17.730 00:30:23.979 Uttam Kumaran: what we, what we did is actually just room. We just like, put this into their separate categories.

187 00:30:24.550 00:30:35.490 Uttam Kumaran: And you still have some phone coming in. So let us know. And here’s and let us know what to do like. I’m not a fan of removing data like, I’m more of a fan of categorize it somewhere else.

188 00:30:35.670 00:30:40.899 Uttam Kumaran: If we remove rows, it’s gonna come. Someone’s gonna come back and say, we wanted the phone some point

189 00:30:41.430 00:30:48.600 Caio Velasco: So in case you want to check, he just replied, a month. So he said, Yeah, those are the missed calls.

190 00:30:51.790 00:30:53.300 Caio Velasco: I can still send

191 00:30:54.270 00:31:04.559 Caio Velasco: cause. I mean, I think the the point is like, what was the question in the beginning? Because when I started, I said, like Robert, mentioned that Javi turned off the phone lines.

192 00:31:05.520 00:31:11.953 Caio Velasco: Then he said, just now, yes, I mentioned that all the phone tickets just say missed a call

193 00:31:12.230 00:31:19.109 Uttam Kumaran: You mentioned. Yeah. So then, so that’s the thing is like, if they if they, if they don’t want us to have it, then maybe we

194 00:31:19.490 00:31:22.580 Uttam Kumaran: we have to filter out missed call phones

195 00:31:24.050 00:31:30.029 Uttam Kumaran: right? But, like again, I don’t. I just want them to tell us what to do here, cause it’s on. I want to put the blame on them

196 00:31:30.190 00:31:32.310 Uttam Kumaran: if they tell us to filter it out and then it

197 00:31:32.830 00:31:36.639 Uttam Kumaran: somewhere. Someone has a phone and it’s still working. And they’re like, where is this?

198 00:31:36.800 00:31:38.990 Uttam Kumaran: 6 months later it’ll happen like this.

199 00:31:39.980 00:31:40.910 Caio Velasco: It’s Hilary.

200 00:31:44.560 00:31:50.279 Caio Velasco: and okay. But what we created the dashboard right? Or it was it on Monday?

201 00:31:54.310 00:31:55.369 Caio Velasco: Because this is still

202 00:31:55.990 00:32:05.739 Caio Velasco: so? Did we create the the we created the charge right? So at the end of the day was on our fault. I guess that we didn’t put the

203 00:32:05.740 00:32:22.170 Uttam Kumaran: No, it’s our. It’s our fault that we package it into other. But what should have happened like ideally? What should have happened is in that call they should have gone and said, What’s in other? Oh, it’s phone and Tiktok shop. Oh, we’re not running the phone anymore. Why is it there. Okay, well, people are calling the phone.

204 00:32:22.750 00:32:27.710 Uttam Kumaran: So maybe you guys should figure that out like, right.

205 00:32:28.000 00:32:31.460 Uttam Kumaran: this is what I’m saying. Not half the problems are not on us.

206 00:32:31.740 00:32:35.769 Uttam Kumaran: That’s the thing. On all these, on everything in data world.

207 00:32:35.970 00:32:41.359 Uttam Kumaran: half the problems are not modeling changes. This is happening today in in Eden. Can I show you an example?

208 00:32:41.550 00:32:42.330 Caio Velasco: Yes.

209 00:32:42.330 00:32:45.259 Uttam Kumaran: I love this topic because this is like story of my life.

210 00:32:45.470 00:32:54.130 Uttam Kumaran: Okay, so on. Eden, the team is working on bringing data into tableau.

211 00:32:54.770 00:32:55.640 Uttam Kumaran: Alright!

212 00:32:56.050 00:33:06.150 Uttam Kumaran: Of course we’ve had. We have them. We have. We’ve had these tables here for like 2 weeks now, just now they’re getting around to putting it into tableau. And then there’s a question saying.

213 00:33:06.300 00:33:11.800 Uttam Kumaran: do we have a way to aggregate the tickets by channel pharmacy? I’m not seeing those fields, in fact tickets

214 00:33:12.230 00:33:16.639 Uttam Kumaran: I wake up this morning. Well, I don’t. This one came in after my meeting. I was like

215 00:33:17.050 00:33:24.969 Uttam Kumaran: one number one. I know oas worked on this 100. This data is in there. I reviewed the Pr, I know it’s there. So point number one is.

216 00:33:26.220 00:33:28.389 Uttam Kumaran: this is not right. It’s not accurate.

217 00:33:28.610 00:33:32.550 Uttam Kumaran: right. But again it’s fine. So so then await responds

218 00:33:33.370 00:33:39.030 Uttam Kumaran: cool fact tickets has customer Id and order number which could be joined to to get pharmacy.

219 00:33:40.210 00:33:46.820 Uttam Kumaran: and then away goes ahead and creates a new table. But this table is literally just a 1 line joint.

220 00:33:47.870 00:33:49.740 Uttam Kumaran: You can do this in tableau

221 00:33:50.120 00:33:50.780 Caio Velasco: Yeah.

222 00:33:50.780 00:33:55.289 Uttam Kumaran: Right. There’s no need for us to be doing this in them. I mean, we could but

223 00:33:56.330 00:33:59.830 Uttam Kumaran: like Look, it’s literally a join just to bring one field in

224 00:34:01.040 00:34:03.519 Uttam Kumaran: complete waste of a table. My mind

225 00:34:03.710 00:34:12.430 Uttam Kumaran: I this is I don’t. I don’t like. I don’t see a need for this, and then he’s like, and then it’s like, Yeah, why, this seems pretty simple. I’m like, do this in tableau.

226 00:34:12.830 00:34:20.960 Uttam Kumaran: I’m like, you don’t need a table. I’m like, that’s what I’m saying, you know. And it’s like.

227 00:34:21.170 00:34:23.339 Uttam Kumaran: just join this. And it’s like, cool.

228 00:34:24.670 00:34:28.539 Uttam Kumaran: Yeah, right? So this could have been something way bigger and a wish

229 00:34:28.710 00:34:31.540 Uttam Kumaran: wasted time doing this for no reason

230 00:34:31.960 00:34:32.600 Caio Velasco: Yeah.

231 00:34:32.840 00:34:38.630 Uttam Kumaran: Perfect example of customer does not know what they want. It’s like, you know. Have you heard that Steve Jobs quote?

232 00:34:39.485 00:34:44.039 Uttam Kumaran: He says customer doesn’t know what they want very famously.

233 00:34:44.360 00:34:46.326 Uttam Kumaran: Give the he-, he said.

234 00:34:46.960 00:34:52.976 Uttam Kumaran: it’s not the customer’s job to know what they want right very famously. That’s how I feel

235 00:34:53.530 00:34:59.079 Uttam Kumaran: they have a problem. Right? So Hana’s problem is probably that she looked at it for 2 seconds and was like.

236 00:34:59.260 00:35:00.740 Uttam Kumaran: I don’t know how to do this drawing.

237 00:35:01.250 00:35:02.959 Uttam Kumaran: you should go make another table.

238 00:35:03.580 00:35:06.609 Uttam Kumaran: Our thing is that no, here’s the joy

239 00:35:08.000 00:35:16.100 Uttam Kumaran: done. End of story, you know. This could have been another 2 h of work back and forth, back and forth, right? It’s useless. It’s

240 00:35:16.410 00:35:18.690 Uttam Kumaran: so. We always have to be skeptical on our side.

241 00:35:19.190 00:35:21.720 Uttam Kumaran: On our side we always have to be skeptical, like.

242 00:35:21.960 00:35:25.366 Uttam Kumaran: you know, we can even go in something recently on Javi, like,

243 00:35:29.940 00:35:33.179 Uttam Kumaran: just use product type from shopify per funnel.

244 00:35:34.850 00:35:40.490 Uttam Kumaran: Yeah, I don’t know what this is include refunds plus every cogs split header titles. These are.

245 00:35:42.090 00:35:44.799 Uttam Kumaran: these are all graphics. Right?

246 00:35:45.340 00:35:47.550 Uttam Kumaran: So okay, let’s how about we take one of these

247 00:35:48.970 00:35:51.417 Uttam Kumaran: and let’s work. Let’s work on it together. So

248 00:35:52.340 00:35:54.489 Caio Velasco: Yeah, let me just verify one.

249 00:35:54.820 00:35:55.800 Caio Velasco: I’m muted

250 00:35:55.800 00:35:56.919 Uttam Kumaran: What’s a good one?

251 00:35:58.650 00:36:02.339 Uttam Kumaran: Just use the product type from shopify per funnel.

252 00:36:05.840 00:36:07.400 Uttam Kumaran: Let’s should we take this one

253 00:36:08.994 00:36:15.769 Caio Velasco: Yeah, I haven’t touched anything related to our, but we can start with that. And I’m going to just reply that month, and

254 00:36:15.770 00:36:17.639 Uttam Kumaran: Okay, hey? Reply to, yeah, get that out

255 00:36:18.610 00:36:18.955 Caio Velasco: Yeah.

256 00:37:04.110 00:37:06.297 Caio Velasco: So just to get back to this

257 00:37:07.580 00:37:13.580 Caio Velasco: So since we open order and we saw phone over there in the last whatever 50 days

258 00:37:14.750 00:37:17.680 Caio Velasco: Then the question for a month should be.

259 00:37:17.840 00:37:20.649 Caio Velasco: should we filter this in the dashboard

260 00:37:21.020 00:37:24.780 Caio Velasco: point on our side, or not? Even that

261 00:37:25.190 00:37:26.230 Uttam Kumaran: No idea.

262 00:37:26.230 00:37:30.300 Uttam Kumaran: Well, I think we should start 1st and look at the dashboard right?

263 00:37:31.120 00:37:34.289 Uttam Kumaran: So if we go look at like

264 00:37:37.290 00:37:40.750 Caio Velasco: Yeah, because the easiest for sure is just to do it there. Yeah.

265 00:37:41.330 00:37:44.790 Uttam Kumaran: So if we go to gross margin dashboard, and there’s like some sort of

266 00:37:49.930 00:37:52.020 Uttam Kumaran: product category.

267 00:37:53.710 00:37:55.389 Uttam Kumaran: So what was his question?

268 00:38:17.330 00:38:21.759 Uttam Kumaran: Let me see if there’s any other conversations about type. Because I feel like I talked about this.

269 00:39:04.370 00:39:07.639 Uttam Kumaran: Okay, so here’s actually the 1st time where I

270 00:39:11.810 00:39:18.889 Uttam Kumaran: client does not like the way we structure product categories. Why do we not use these product type? What would it like to be to hide accessories?

271 00:39:19.240 00:39:21.749 Uttam Kumaran: They only care about protein conference rate.

272 00:39:22.740 00:39:28.270 Uttam Kumaran: We worked with them on this. Many of the product fields are improperly tagged, so we filled it out with them on.

273 00:39:30.290 00:39:33.050 Uttam Kumaran: So probably what he’s asking here

274 00:39:33.340 00:39:36.359 Uttam Kumaran: is just to revert back to using product type.

275 00:39:36.610 00:39:40.969 Uttam Kumaran: One of the things that we can just try here really quickly. And to see what it looks like

276 00:39:42.960 00:39:47.979 Uttam Kumaran: is just going into here and saying, cool. I want to edit this.

277 00:39:48.450 00:39:53.610 Uttam Kumaran: and I want to bring in the product type here.

278 00:39:54.130 00:39:59.149 Uttam Kumaran: set of category. And one of the things that before I play on this.

279 00:39:59.320 00:40:01.619 Uttam Kumaran: I’m just gonna go into Github and look at

280 00:40:01.900 00:40:03.950 Uttam Kumaran: what? What is this product type?

281 00:40:05.133 00:40:06.099 Uttam Kumaran: Coming from.

282 00:40:06.500 00:40:11.761 Uttam Kumaran: So if I go into code, and this is pulling in from

283 00:40:14.170 00:40:19.079 Uttam Kumaran: this is pulling it from fact order line right? So if I go in here. And I find fact

284 00:40:19.400 00:40:20.690 Uttam Kumaran: borderline.

285 00:40:22.430 00:40:25.589 Uttam Kumaran: I’m gonna go look for product type, right?

286 00:40:26.750 00:40:31.609 Uttam Kumaran: So there’s nothing coming in for Amazon as product type.

287 00:40:36.570 00:40:41.989 Uttam Kumaran: But it looks like order line. That product type is coming from into shopify order line.

288 00:40:43.030 00:40:44.920 Uttam Kumaran: So let’s go look at that. Now

289 00:40:49.110 00:40:53.090 Uttam Kumaran: in here. Let’s see if we have product type.

290 00:40:54.450 00:40:58.969 Uttam Kumaran: Okay, so there is a product type field. It looks like this is coming directly from

291 00:40:59.610 00:41:06.779 Uttam Kumaran: P dot product type, which is directly from the source or in shopify product. I guess we can check one more

292 00:41:07.300 00:41:09.359 Uttam Kumaran: and stop the phone.

293 00:41:12.720 00:41:15.609 Uttam Kumaran: Okay, coming directly from the source.

294 00:41:16.210 00:41:23.719 Uttam Kumaran: So if if his thing like, we’re from reading that I already gave some feedback that the data is not gonna look good.

295 00:41:25.560 00:41:30.150 Uttam Kumaran: The pushback was, who cares? So I’m gonna go ahead and say.

296 00:41:30.840 00:41:41.150 Uttam Kumaran: I’m gonna go ahead and make the change. We’re gonna visualize it. Let’s go back to just this.

297 00:41:41.660 00:41:45.039 Uttam Kumaran: Just this one. But edit this question

298 00:41:45.580 00:41:48.280 Uttam Kumaran: and move out product category. And I add in

299 00:41:48.520 00:41:52.360 Uttam Kumaran: product type, and I put it in the middle here again.

300 00:41:52.860 00:41:56.450 Uttam Kumaran: and I want to hit, save and replace.

301 00:42:00.860 00:42:03.529 Uttam Kumaran: I don’t see any of the product categories. So here

302 00:42:07.140 00:42:08.340 Uttam Kumaran: or product type.

303 00:42:14.710 00:42:15.630 Uttam Kumaran: okay.

304 00:42:33.340 00:42:36.530 Uttam Kumaran: product, type, customer type.

305 00:42:41.510 00:42:42.810 Uttam Kumaran: that’ll look fine.

306 00:43:12.470 00:43:16.260 Uttam Kumaran: Yeah, I guess my question is, why isn’t this?

307 00:43:35.140 00:43:36.060 Uttam Kumaran: Oh.

308 00:43:44.030 00:43:45.930 Uttam Kumaran: so now if I hit save.

309 00:43:58.080 00:43:58.770 Uttam Kumaran: Okay.

310 00:44:01.270 00:44:04.600 Uttam Kumaran: So to give you a sense of what happened here.

311 00:44:04.940 00:44:10.089 Uttam Kumaran: we built a product categorization mapping file with Aman.

312 00:44:10.410 00:44:13.799 Uttam Kumaran: When we presented to the CEO. They didn’t like it.

313 00:44:16.650 00:44:18.719 Uttam Kumaran: So that’s it about it like.

314 00:44:18.920 00:44:22.389 Uttam Kumaran: that’s it, that’s all I know. And right see, I even asked. Here

315 00:44:22.610 00:44:32.039 Uttam Kumaran: we worked on with them on on this, what should we do? And the the response was.

316 00:44:32.330 00:44:35.380 Uttam Kumaran: just use product type from shopify.

317 00:44:38.180 00:44:39.310 Uttam Kumaran: So

318 00:44:40.390 00:44:45.779 Uttam Kumaran: that that’s the change we need to make is basically go through anytime where product category is being used here.

319 00:44:45.990 00:44:47.600 Uttam Kumaran: we can switch it to product type.

320 00:44:55.470 00:44:56.220 Uttam Kumaran: So

321 00:44:57.340 00:44:57.910 Awaish Kumar: Right.

322 00:44:58.120 00:45:02.249 Awaish Kumar: What was the difference like? These are the same values in both the

323 00:45:04.370 00:45:12.779 Uttam Kumaran: Yeah, the difference. The difference is like, I don’t allow for anything that’s uncategorized and

324 00:45:13.960 00:45:15.949 Uttam Kumaran: like, if we go to the

325 00:45:21.190 00:45:31.120 Uttam Kumaran: the data platform documentation. This is what I built with him, which is this. These were all categorized as as null like not having a type.

326 00:45:32.600 00:45:38.790 Uttam Kumaran: So then I worked with him on categorizing these as other stuff, right? Like.

327 00:45:39.660 00:45:44.780 Uttam Kumaran: for example. So they didn’t like the categorization fine, whatever like. That’s

328 00:45:46.450 00:45:53.240 Uttam Kumaran: it’s up to them to debate which one they wanted to use. I had a discussion with them on, he said, go with this. The CEO doesn’t want that. So

329 00:45:53.870 00:45:55.210 Uttam Kumaran: that’s fine, you know.

330 00:45:58.470 00:46:04.930 Uttam Kumaran: we still have the product category. But so

331 00:46:13.030 00:46:13.830 Uttam Kumaran: okay,

332 00:46:17.740 00:46:21.960 Uttam Kumaran: So I think, probably this is the only other one I mean, I could finish up this this fix.

333 00:46:22.730 00:46:24.219 Uttam Kumaran: Do you want to try to take?

334 00:46:25.740 00:46:27.440 Uttam Kumaran: You want to try to take this?

335 00:46:29.480 00:46:33.000 Uttam Kumaran: So it looks like another cogs. Category has to be added.

336 00:46:33.210 00:46:39.360 Uttam Kumaran: which is 2.9% plus 30 cents on every order.

337 00:46:45.690 00:46:52.380 Uttam Kumaran: Basically what he’s asking for here is every order we have to pay a shopify fee.

338 00:46:52.970 00:46:54.729 Uttam Kumaran: But that doesn’t come in.

339 00:46:55.310 00:46:58.000 Uttam Kumaran: That’s not coming in in the data that’s coming in like

340 00:46:58.170 00:47:03.000 Uttam Kumaran: they’re getting billed by shopify. So we want to consider that as part of cost of goods sold.

341 00:47:05.160 00:47:09.440 Luke Daque: Maybe, and hard code that for shopify. Also that’d be fine

342 00:47:09.780 00:47:11.550 Uttam Kumaran: It’s probably a hard code, and

343 00:47:11.690 00:47:13.639 Uttam Kumaran: and one of the shopify order

344 00:47:15.080 00:47:15.570 Luke Daque: Yeah.

345 00:47:15.570 00:47:16.010 Uttam Kumaran: Tables.

346 00:47:16.010 00:47:21.490 Luke Daque: Sucks that shopify doesn’t have that Amazon does have their piece in a table

347 00:47:21.960 00:47:22.570 Uttam Kumaran: Yeah.

348 00:47:33.150 00:47:36.849 Uttam Kumaran: So that’s those 2. So I mean, I’m just, I’ll finish up this product type.

349 00:47:37.530 00:47:40.350 Uttam Kumaran: I don’t know. Kai, are you off? You’re gonna be off soon.

350 00:47:40.850 00:47:41.780 Uttam Kumaran: Probably

351 00:47:42.698 00:47:44.249 Caio Velasco: More or less? Yes.

352 00:47:44.250 00:47:45.460 Uttam Kumaran: Okay. Okay. Then I’ll then

353 00:47:45.460 00:47:46.109 Caio Velasco: Do need

354 00:47:46.730 00:47:50.119 Uttam Kumaran: We’ll take. Then we’ll take. We’ll just finish up the other fee.

355 00:47:50.540 00:47:55.659 Uttam Kumaran: But I guess long story short. Yeah, I mean, this is like sort of how I’m just triaging everything and going. So

356 00:47:56.070 00:48:00.200 Uttam Kumaran: I think soon, I I looked in linear and some stuff is getting added there.

357 00:48:00.370 00:48:04.509 Uttam Kumaran: So I’m gonna make sure that both these fixes get. I’ll update the linear ticket

358 00:48:04.810 00:48:06.179 Uttam Kumaran: with how I did it.

359 00:48:08.170 00:48:12.250 Uttam Kumaran: And then that’ll that’ll basically allow us to push the gross margin fixes out

360 00:48:14.820 00:48:23.849 Caio Velasco: Okay, yeah, for this. Well, for this part like this, this work here, like I, I, at least in the beginning, I do feel that I need time to

361 00:48:23.850 00:48:24.270 Uttam Kumaran: Yeah.

362 00:48:24.544 00:48:30.039 Caio Velasco: Understand. I have no idea what cogs is. And let’s let’s let’s assume that I’m a newbie in this part

363 00:48:30.040 00:48:30.710 Uttam Kumaran: Yeah.

364 00:48:31.409 00:48:35.459 Caio Velasco: But then I’ll spend time as we may. As as we talked, learning this

365 00:48:35.460 00:48:38.869 Uttam Kumaran: Take time, and let’s try to do as many of these sessions as we can

366 00:48:39.380 00:48:40.050 Caio Velasco: Okay.

367 00:48:40.050 00:48:41.470 Uttam Kumaran: Get through stuff quick.

368 00:48:41.620 00:48:46.639 Uttam Kumaran: I think once you watch me and watch some of the other folks, and as we’re sort of sharing the knowledge

369 00:48:48.200 00:48:50.089 Caio Velasco: So it’s quite helpful.

370 00:48:50.370 00:48:59.010 Caio Velasco: And if you can, if you can like on the data team channel, I just based on

371 00:49:00.391 00:49:08.169 Caio Velasco: which one the client. Javi, yeah, on the on the screenshot you send.

372 00:49:08.850 00:49:17.250 Caio Velasco: I pasted what I’m on, replied I just wanted to to quickly check it, just to see like if I understand what is happening at the end of the day.

373 00:49:17.440 00:49:25.409 Caio Velasco: because from his answers, then I don’t know if we

374 00:49:27.560 00:49:36.049 Caio Velasco: I don’t know I’m I’m still a bit lost like why would would he say like just filter it? So he’s not looking even at the dashboard like not even

375 00:49:38.270 00:49:41.849 Uttam Kumaran: I guess. I know I mean, but that’s what I’m saying. He doesn’t know.

376 00:49:43.390 00:49:44.520 Uttam Kumaran: That’s what. Just ask.

377 00:49:44.630 00:49:47.659 Uttam Kumaran: Just ask the question, did you send it, just just send it to him.

378 00:49:52.990 00:50:02.956 Caio Velasco: Anyway, I sent him that I. We just opened the data within order. And then phone is still there, because exactly what he said like that dismissive calls whatever.

379 00:50:03.827 00:50:10.269 Caio Velasco: And then I asked, like, can we just filter out phone and SMS directly in the chart? And he’s like, I’m not sure what you mean.

380 00:50:10.420 00:50:11.769 Caio Velasco: Just hide them.

381 00:50:14.390 00:50:17.030 Caio Velasco: Okay. So whoever built the dashboard then

382 00:50:17.160 00:50:21.789 Uttam Kumaran: That’s why that that’s at the end of the what we need to do. Like. Just.

383 00:50:21.790 00:50:25.830 Uttam Kumaran: he said, oh, I see your thing right now, right? Okay, let me pull it up on my side.

384 00:50:26.050 00:50:33.449 Uttam Kumaran: Not sure what you mean. We don’t need SMS appearing any charts. Those tickets don’t exist on so you can keep them in dB and hide them from all charts if that works.

385 00:50:35.870 00:50:40.860 Uttam Kumaran: Okay? So we asked once, let me ask twice. That’s it done.

386 00:50:41.180 00:50:45.710 Uttam Kumaran: So what we’ll do is I’m just gonna we’ll just go in here

387 00:50:45.870 00:50:51.710 Uttam Kumaran: and then just filter those out, which is fine.

388 00:50:51.940 00:50:55.209 Uttam Kumaran: Yeah, just need to know that that’s what they want. Right?

389 00:50:57.350 00:50:58.379 Uttam Kumaran: They’ll go and

390 00:50:58.380 00:51:02.800 Caio Velasco: And then, and then we assumed that later one day they was like, Hey, I added, stuff

391 00:51:02.800 00:51:03.460 Uttam Kumaran: But like I said

392 00:51:03.460 00:51:04.679 Caio Velasco: Why is not configured

393 00:51:04.680 00:51:11.289 Uttam Kumaran: Yes, but ultimately, like you can, you can lead the horse to the water, but you can’t say drink right? So

394 00:51:11.410 00:51:13.600 Uttam Kumaran: that’s the problem.

395 00:51:14.800 00:51:17.102 Uttam Kumaran: It is what it is, what it is.

396 00:51:17.840 00:51:22.350 Uttam Kumaran: So I’m gonna go, is not. And I’m gonna say, is not SMS and is not phone.

397 00:51:22.760 00:51:24.180 Uttam Kumaran: Great fun.

398 00:51:25.080 00:51:25.800 Uttam Kumaran: Thanks.

399 00:51:26.510 00:51:29.653 Uttam Kumaran: Solid mission accomplished

400 00:51:30.340 00:51:30.790 Caio Velasco: Done

401 00:51:31.240 00:51:36.970 Uttam Kumaran: So can you send? Can you send to Kyle? Can you send? Can you send to Aman to say cool changes made done

402 00:51:38.050 00:51:39.205 Uttam Kumaran: great

403 00:51:41.070 00:51:48.170 Uttam Kumaran: part of part of this job is this comical so like, but that’s the thing. It’s every client. This is all new to them. They don’t know what we know

404 00:51:49.400 00:51:52.220 Luke Daque: Yeah. And sometimes we just overthink, I guess

405 00:51:52.370 00:51:58.020 Uttam Kumaran: Yeah. So we’re engineers. So we’re gonna think that they know so much. And if they don’t know sometimes. So it’s fine, whatever

406 00:52:01.680 00:52:09.720 Luke Daque: Like, sometimes it’s just a basic thing like that. And then you, I’ll be thinking, like, maybe I need to change the data model or something. And then

407 00:52:10.700 00:52:18.069 Luke Daque: I think that’s what a wish did earlier, right? Like, he just created a different model just with that specific filter or something

408 00:52:40.760 00:52:53.780 Caio Velasco: Okay. So okay. So the only thing they wanted from what Robert asks is to hide it, that just a dashboard fixed call. I would never expect that.

409 00:52:54.810 00:52:55.550 Caio Velasco: Okay.

410 00:52:55.550 00:52:56.939 Uttam Kumaran: But that’s what I’m saying. This is the thing

411 00:52:56.940 00:52:58.800 Caio Velasco: Lesson, learned thanks.

412 00:52:59.090 00:53:06.889 Uttam Kumaran: Yeah, it’s just you have to chase it. And your goal is to assume your assumption. And Mike will talk to Mike about this when you talk to him

413 00:53:07.080 00:53:10.829 Uttam Kumaran: you should assume that 50% of things are not problems.

414 00:53:11.110 00:53:13.880 Uttam Kumaran: But this is like what a Pm. Is supposed to do right.

415 00:53:14.120 00:53:16.320 Uttam Kumaran: And I know this because I was a Pm.

416 00:53:16.810 00:53:26.259 Uttam Kumaran: For 2 years, like for Pm and data, everything is gonna come. And your job is to make sure 50% of those are non problems, or someone can go change really quickly.

417 00:53:26.430 00:53:32.440 Uttam Kumaran: Right? This should have been like, oh, so like someone the Pm should have or an analyst should have figured this out and made the change.

418 00:53:32.990 00:53:36.269 Uttam Kumaran: That’s it. It. This happens so

419 00:53:48.990 00:53:53.770 Uttam Kumaran: okay, I guess I wanted to spend a little bit of time. Who do I have? Who’s on this call with us?

420 00:53:54.250 00:53:59.389 Uttam Kumaran: Yeah, okay, I guess I mean, Kyle, you’re free. You’re free to drop. But I guess I wanted to spend a little bit of time talking about

421 00:53:59.740 00:54:05.663 Uttam Kumaran: by my thesis for how I’m gonna do this documentation.

422 00:54:06.958 00:54:11.919 Uttam Kumaran: If you want to say you can listen, or I can. We’ll record this. I can send it tomorrow, too.

423 00:54:13.730 00:54:19.380 Caio Velasco: Okay, cool. I can. Actually, I’m I can grab something to eat. And I can say, for like a bit more

424 00:54:19.640 00:54:22.080 Uttam Kumaran: That’s fine. Do- do whatever do whatever

425 00:54:22.080 00:54:24.920 Caio Velasco: Okay, okay, I’ll be right back. But if you okay, perfect

426 00:54:31.160 00:54:32.080 Uttam Kumaran: Okay.

427 00:54:36.330 00:54:40.229 Uttam Kumaran: Well, let me just finish up a couple. Let me just close out a couple more things.

428 00:55:13.810 00:55:15.530 Uttam Kumaran: Let me invite you.

429 00:55:16.310 00:55:17.280 Uttam Kumaran: Oh, aish!

430 00:55:17.590 00:55:18.420 Uttam Kumaran: Right now.

431 00:55:49.160 00:55:53.729 Caio Velasco: Hey? I do have to drop but I’ll I’ll make sure to watch it

432 00:55:53.730 00:55:54.240 Uttam Kumaran: All good.

433 00:55:54.310 00:55:55.230 Caio Velasco: 2 more.

434 00:55:55.520 00:55:56.649 Uttam Kumaran: Yeah. No problem.

435 00:55:57.120 00:55:57.980 Uttam Kumaran: See? Ya.

436 00:55:59.230 00:55:59.970 Caio Velasco: See ya

437 00:57:33.850 00:57:42.630 Luke Daque: Bhutan for stack Blitz stack Blitz the lifetime value. Ltv.

438 00:57:43.080 00:57:50.179 Luke Daque: Does it make sense to be the sum of subscriptions, revenue and token reloads?

439 00:57:50.740 00:57:54.560 Luke Daque: Or should we split it into to

440 00:57:55.863 00:58:00.389 Uttam Kumaran: I think it’s probably worth having both, and then doing one. That’s a sum.

441 00:58:00.650 00:58:03.700 Uttam Kumaran: Can you send it? Can you ask Mitch just to confirm

442 00:58:05.450 00:58:06.450 Luke Daque: Yeah, sure.

443 00:58:08.750 00:58:15.510 Luke Daque: But yeah, I think it’s yeah. Maybe I’ll just say like, it would be great if we can filter

444 00:58:16.180 00:58:19.560 Luke Daque: both right. So maybe just show them everything.

445 00:58:19.790 00:58:22.919 Luke Daque: There’s a sum, and then it can be filtered. So yeah.

446 01:02:02.820 01:02:03.930 Uttam Kumaran: Away, is she there?

447 01:02:06.680 01:02:08.291 Uttam Kumaran: Okay, I’m gonna tag you on

448 01:02:08.980 01:02:12.670 Uttam Kumaran: on 2 tickets for Javi. Do you have a moment?

449 01:02:14.340 01:02:17.440 Uttam Kumaran: To take these. This would be very, very helpful.

450 01:02:20.130 01:02:23.930 Uttam Kumaran: This one and

451 01:06:08.970 01:06:11.109 Awaish Kumar: Are we having that conversation now?

452 01:06:14.240 01:06:15.580 Awaish Kumar: Like we had a meeting.

453 01:06:16.480 01:06:19.719 Awaish Kumar: Oh, right now, if

454 01:06:20.130 01:06:30.950 Uttam Kumaran: Oh, yes, no, I have. I have. I have that. I want to go through a couple of items. I’m just finding I I if I just pick ping you in Javi if you can. Can you take this ticket today?

455 01:06:32.790 01:06:36.399 Uttam Kumaran: And then I’m just I just wanna find if there’s any other stuff I can

456 01:06:40.570 01:06:47.560 Awaish Kumar: So is this like, I just have to add this as a like. For all the orders.

457 01:06:48.080 01:06:54.349 Awaish Kumar: we get 2.9% of what total order, or like total price of what?

458 01:06:55.279 01:06:56.869 Uttam Kumaran: Let me confirm.

459 01:07:25.550 01:07:28.040 Uttam Kumaran: So it’s the total transaction price

460 01:07:31.550 01:07:33.599 Awaish Kumar: The total price of the order. Right?

461 01:07:33.800 01:07:34.740 Awaish Kumar: Yeah? So

462 01:07:35.266 01:07:36.319 Uttam Kumaran: After discounts.

463 01:07:40.660 01:07:46.170 Awaish Kumar: Yeah, okay, pretax after discount. Right?

464 01:07:47.120 01:07:50.639 Uttam Kumaran: Correct and let me. Yeah, I’m just gonna confirm

465 01:09:42.109 01:09:45.979 Awaish Kumar: Should rename it as shopify payment, cogs

466 01:09:48.240 01:09:49.040 Uttam Kumaran: Hmm.

467 01:09:51.479 01:09:55.159 Awaish Kumar: The name for the field should be like like shopify payment code.

468 01:09:55.160 01:09:58.269 Uttam Kumaran: You can do shop. You can do shopify platform cogs.

469 01:10:00.600 01:10:01.190 Awaish Kumar: Okay.

470 01:18:23.907 01:18:27.279 Uttam Kumaran: I’m gonna just spend a second, maybe. Let’s talk about

471 01:18:28.330 01:18:31.453 Uttam Kumaran: what I’m thinking about for documentation. And

472 01:19:00.650 01:19:03.510 Uttam Kumaran: you guys have like, 10 min. If you guys just wanna

473 01:19:03.840 01:19:07.540 Uttam Kumaran: follow me for a sec, I just wanna get your approval on this

474 01:19:09.210 01:19:12.199 Uttam Kumaran: so basically, what I’m thinking is,

475 01:19:15.230 01:19:19.219 Uttam Kumaran: we have this like Pm. Handoff, doc.

476 01:19:20.221 01:19:24.889 Uttam Kumaran: So I’m kind of start. I’m kind of thinking of this as 3 big rocks.

477 01:19:25.769 01:19:28.619 Uttam Kumaran: I’m thinking of engagement overview.

478 01:19:30.072 01:19:31.360 Uttam Kumaran: Give me one second.

479 01:19:35.530 01:19:40.509 Uttam Kumaran: So I’m kind of thinking about. This is like having one thing which is like the engagement or overview.

480 01:19:41.490 01:19:44.079 Uttam Kumaran: I’m also thinking of this having the

481 01:19:45.290 01:19:46.600 Luke Daque: Priorities.

482 01:19:47.180 01:19:53.570 Uttam Kumaran: Work streams, outcomes and impacts.

483 01:20:01.710 01:20:09.920 Uttam Kumaran: So actually, probably this is not necessary question, jeez

484 01:20:12.550 01:20:19.320 Uttam Kumaran: outcomes and impacts. Then we’re gonna have a implementation overview.

485 01:20:25.600 01:20:31.790 Uttam Kumaran: Yeah. So I think these are the, these are the 4 kind of core

486 01:20:32.120 01:20:35.909 Uttam Kumaran: sections I’m thinking about. And let me walk through what’s gonna be in each of these.

487 01:20:36.130 01:20:40.069 Uttam Kumaran: So the overall structure of this document

488 01:20:44.330 01:20:46.520 Uttam Kumaran: is gonna be FAQ, based.

489 01:20:48.900 01:21:01.520 Uttam Kumaran: meaning it’s all gonna be surrounded around question. Answer question. Answer the reason. I think it’s gonna be helpful to do that is one. I want this document to be able to be used by AI really easily.

490 01:21:02.302 01:21:06.950 Uttam Kumaran: And I think that a lot of the times people have questions and we want them to have the answer.

491 01:21:07.140 01:21:11.169 Uttam Kumaran: We do if they want to organize these questions, though. So there’s also gonna be domains.

492 01:21:11.280 01:21:14.139 Uttam Kumaran: Domains is gonna be things like shipping.

493 01:21:15.610 01:21:17.110 Uttam Kumaran: Maybe cogs.

494 01:21:18.115 01:21:22.909 Uttam Kumaran: Maybe customer service, right? These are going to be domain specific.

495 01:21:23.290 01:21:28.999 Uttam Kumaran: There’s also gonna be things around like the platform

496 01:21:29.210 01:21:37.010 Uttam Kumaran: which is gonna be like ingestion, which is gonna be modeling, which is gonna be storage.

497 01:21:37.740 01:21:38.800 Uttam Kumaran: And this

498 01:21:39.530 01:21:45.350 Uttam Kumaran: right. That’s probably most of it. So really the engagement overview. I see it as all of these questions.

499 01:21:47.300 01:21:51.930 Uttam Kumaran: and Akash provided us with a great overview here of of these.

500 01:21:52.860 01:21:59.580 Uttam Kumaran: but I see it as looking through these 4 areas.

501 01:22:00.354 01:22:07.540 Uttam Kumaran: client and account history sales, history, client and company, industry, history, org, chart, past projects

502 01:22:07.700 01:22:10.810 Uttam Kumaran: current project, ramp up deliverables.

503 01:22:11.250 01:22:18.600 Uttam Kumaran: And then this one I’m going to. This is all gonna be sort of like business focused things

504 01:22:19.210 01:22:29.179 Uttam Kumaran: right? So then I’m then there’s a couple so kind of the top part here is all gonna be around the engagement, the second part here work streams and priorities. This is gonna be like

505 01:22:29.420 01:22:34.880 Uttam Kumaran: this is gonna be like work stream. This is gonna be like a for example.

506 01:22:35.780 01:22:44.240 Uttam Kumaran: If I do reduce shipping costs right?

507 01:22:44.380 01:22:47.709 Uttam Kumaran: This is. Gonna and then we’re gonna have like priority.

508 01:22:48.040 01:23:07.700 Uttam Kumaran: I, this is gonna be like just information about trying to renegotiate shipping cost right or gain insight into

509 01:23:08.130 01:23:11.400 Uttam Kumaran: profitability by product.

510 01:23:11.690 01:23:22.230 Uttam Kumaran: This is gonna be priority medium. This is gonna mean need to break out products and find individual

511 01:23:22.930 01:23:35.780 Uttam Kumaran: product level profitability profit equals revenue minus discounts minus cogs minus 3 folks right

512 01:23:36.120 01:23:40.870 Uttam Kumaran: outcomes and impacts. This is gonna be like the

513 01:23:41.050 01:23:47.159 Uttam Kumaran: date which is gonna be, let’s say, July, this is gonna be like almost like work that we completed for them.

514 01:23:47.870 01:23:51.030 Uttam Kumaran: So dated July 2024,

515 01:23:51.460 01:23:55.878 Uttam Kumaran: the in the work stream is maybe

516 01:23:57.150 01:24:13.020 Uttam Kumaran: gain insight into Amazon business. The impact is gonna be delivered 3 comp comprehensive dashboards and

517 01:24:14.090 01:24:22.180 Uttam Kumaran: allowed for faster. Go to market for new product on Amazon.

518 01:24:23.480 01:24:29.000 Uttam Kumaran: So this is sort of like what’s active ideally, work streams move into outcomes right like

519 01:24:29.370 01:24:36.230 Uttam Kumaran: these should move down here over time. And this is the stuff that we want to be able to turn into case studies

520 01:24:36.430 01:24:38.240 Uttam Kumaran: on the marketing side eventually.

521 01:24:38.370 01:24:46.740 Uttam Kumaran: So sort of things stream down here and then implementation overview. This is where this is everything in our world which is going to be information about the stack.

522 01:24:47.040 01:24:51.550 Uttam Kumaran: It’s gonna be information. So it’s actually 2 things. One, it’s gonna be

523 01:24:52.220 01:24:58.290 Uttam Kumaran: we’re gonna have domain by domain. So we’re gonna talk 1st about like shopify.

524 01:25:00.065 01:25:02.270 Uttam Kumaran: Let’s say, Amazon.

525 01:25:04.855 01:25:09.319 Uttam Kumaran: Let’s say shipping right? So then we’re gonna have faqs about

526 01:25:09.680 01:25:11.830 Uttam Kumaran: an example of thing today could be.

527 01:25:13.670 01:25:18.420 Uttam Kumaran: how do we calculate fees for shopify?

528 01:25:19.800 01:25:30.010 Uttam Kumaran: We have 2.9% and and 30 cents for each transaction.

529 01:25:30.250 01:25:35.350 Uttam Kumaran: See here, in github or lodging.

530 01:25:36.730 01:25:39.529 Uttam Kumaran: So everything’s gonna be oriented like Faqs.

531 01:25:40.250 01:25:46.150 Uttam Kumaran: And we’re gonna start to build up the Faqs as we go. Additionally, I’m gonna use AI

532 01:25:46.290 01:25:48.030 Uttam Kumaran: to kick off

533 01:25:48.210 01:25:55.040 Uttam Kumaran: this document across all of our clients, because AI is not going to be good at doing things in Google sheets

534 01:25:55.170 01:25:59.890 Uttam Kumaran: or things like that. It’s gonna be really good at anything that’s descriptive like this. So

535 01:26:00.120 01:26:05.220 Uttam Kumaran: basically, looking through slack, answering all the top questions and having them available here.

536 01:26:06.240 01:26:10.370 Uttam Kumaran: the one piece that it’s not clear which is like, how do we? So

537 01:26:10.530 01:26:18.980 Uttam Kumaran: also here we could, we could put in like, what dashboards do we have that cover shopify.

538 01:26:20.680 01:26:28.070 Uttam Kumaran: So it’s margin dash link, right? And so I think we sort of start to maintain this.

539 01:26:29.070 01:26:38.540 Uttam Kumaran: I think additionally, we’re we’re gonna probably need to have some sort of like more deeper thing around ingestion

540 01:26:46.800 01:26:50.800 Uttam Kumaran: congestion modeling storage.

541 01:26:51.430 01:26:53.160 Uttam Kumaran: And this right?

542 01:27:01.320 01:27:05.239 Uttam Kumaran: So this is gonna be all related to engineering implementation.

543 01:27:06.090 01:27:10.020 Uttam Kumaran: So this is gonna be around like.

544 01:27:12.590 01:27:19.030 Uttam Kumaran: what’s the difference between dev staging and fraud?

545 01:27:20.430 01:27:27.310 Uttam Kumaran: See our see? Our dbt structure, Doc. Here.

546 01:27:37.040 01:27:45.610 Uttam Kumaran: another thing on ingestion is like, where is X data coming from, it’s in portable.

547 01:27:47.640 01:27:51.589 Uttam Kumaran: We’re gonna be able, once we start to have the programmatic data

548 01:27:52.350 01:28:00.729 Uttam Kumaran: for these, we can start to generate this automatically using AI. But I think the question answer, sort of method is the best way of doing this.

549 01:28:03.180 01:28:04.186 Uttam Kumaran: I think.

550 01:28:06.820 01:28:12.040 Uttam Kumaran: I think there’s there’s honestly like, and then the other. So a couple of other pieces. How this doc works.

551 01:28:13.090 01:28:17.890 Uttam Kumaran: and then in terms of like, how this gets implemented with AI.

552 01:28:18.160 01:28:23.350 Uttam Kumaran: So the AI piece of this is one I’m expecting the AI to have access to Github.

553 01:28:24.020 01:28:32.050 Uttam Kumaran: So it’s like input data. Github slack emails, zooms.

554 01:28:33.460 01:28:36.129 Uttam Kumaran: And the second thing is the question types

555 01:28:37.100 01:28:45.340 Uttam Kumaran: these are gonna be code, lookup, logic question, or who to ask?

556 01:28:46.760 01:28:51.589 Uttam Kumaran: Right? So I’m trying to think about, how do we categorize all questions like any question we have about

557 01:28:52.460 01:28:59.580 Uttam Kumaran: a client. I want to categorize them into into something. So I’ll be working a little bit on this as we figure out the AI piece.

558 01:29:00.470 01:29:02.679 Uttam Kumaran: But what do you guys think of something like this?

559 01:29:05.410 01:29:10.210 Awaish Kumar: Yeah, like, this is a, I think this is a great documentation.

560 01:29:10.450 01:29:16.280 Awaish Kumar: like even for us, like we sometime investigate and answer some questions, and

561 01:29:16.580 01:29:21.940 Awaish Kumar: after a few days we we just forget, like what was what we shared. So

562 01:29:22.160 01:29:24.470 Awaish Kumar: this will be great documentation for

563 01:29:24.640 01:29:28.959 Awaish Kumar: for these kind of investigation investigations. And

564 01:29:29.710 01:29:36.820 Awaish Kumar: like question answers, like the regarding logic regarding business context.

565 01:29:39.030 01:29:40.649 Luke Daque: Yeah, I agree. Like.

566 01:29:41.550 01:30:04.759 Luke Daque: I always forget. I mean, I’m always bad at remembering things. And this would be great. And yeah, like, like Utah mentioned. This would be a great source also, like context for AI that way. If we have, like an agent, we can just ask the agent some questions, and he can look up the Faqs or whatever the the documentation. And yeah, it can answer

567 01:30:05.400 01:30:10.230 Luke Daque: very quickly, as opposed to us, like looking through all the Faqs right

568 01:30:10.230 01:30:13.829 Uttam Kumaran: Well, to tell you where this is gonna go is

569 01:30:16.060 01:30:19.929 Uttam Kumaran: As more stuff happens in zoom slack

570 01:30:20.200 01:30:25.490 Uttam Kumaran: and meetings, the AI is gonna automatically suggest changes

571 01:30:27.250 01:30:33.919 Uttam Kumaran: Meaning. If a meeting happens where we answer 3 more questions and updates happen to 2 existing questions.

572 01:30:34.460 01:30:39.939 Uttam Kumaran: the AI should suggest, Hey, I’m I’m gonna go. Add this question in here.

573 01:30:40.490 01:30:47.930 Uttam Kumaran: So this becomes a dynamic document. Right? What is the what is the core issue with documentation is maintaining it

574 01:30:48.330 01:30:53.830 Uttam Kumaran: with AI. Now, one thing that we built for a customer is, we built this process of updating this

575 01:30:54.100 01:30:55.160 Uttam Kumaran: through AI,

576 01:30:55.300 01:31:04.590 Uttam Kumaran: right? So because this is all stuff that changes over time as these things happen like on a new Pr. The AI will say, do we need to change anything here

577 01:31:05.800 01:31:14.910 Uttam Kumaran: every week, after every week or every few days after we talk in slack, the AI will take all that compared to this and tell what needs to be updated.

578 01:31:15.200 01:31:17.210 Uttam Kumaran: Email zooms and linear.

579 01:31:18.410 01:31:19.879 Uttam Kumaran: Do you see what I mean?

580 01:31:20.080 01:31:21.499 Luke Daque: Yeah, that would be awesome.

581 01:31:21.500 01:31:34.610 Uttam Kumaran: Any question that comes. I want it to be categorized. And I, the other thing is you, everybody. This, this will be going to this, Doc, that find these questions you should be answer asking the AI bot the questions

582 01:31:35.230 01:31:36.000 Luke Daque: Right.

583 01:31:37.330 01:31:38.330 Uttam Kumaran: See what I mean.

584 01:31:39.050 01:31:48.940 Uttam Kumaran: This just becomes the knowledge base. But we’ll be able to access this through the AI and update it through the AI. You can also come in here and find it. But

585 01:31:49.060 01:31:52.359 Uttam Kumaran: this is what I think is gonna probably be probably be best

586 01:31:53.110 01:31:53.890 Luke Daque: Yeah, I agree.

587 01:31:54.160 01:31:57.519 Awaish Kumar: What kind of AI tools will be used for that

588 01:31:58.580 01:32:07.489 Uttam Kumaran: Yeah. So this is something that we’re, we’re probably because of how small this is we probably will use. We probably will use Gemini.

589 01:32:08.600 01:32:11.590 Uttam Kumaran: because all this is, can we can put into context?

590 01:32:12.120 01:32:19.550 Uttam Kumaran: And I think I’ll I’ll do a demo depending on how far I get on this by Friday. I’ll try to do a demo on Friday of how the system works.

591 01:32:23.130 01:32:27.580 Uttam Kumaran: But it’s a lot of it’s partly a data problem, right? We have to move all this stuff

592 01:32:27.860 01:32:30.990 Uttam Kumaran: from the sources into the AI agent.

593 01:32:31.360 01:32:36.009 Uttam Kumaran: But I’m gonna right now, I’m gonna I’m gonna work on building this for Javi today.

594 01:32:36.620 01:32:40.560 Uttam Kumaran: And I’m gonna take in all this data and shove it into context and see what happens

595 01:32:43.060 01:32:49.050 Awaish Kumar: Okay, yeah. My question was more like, how we are going to automate it. Because

596 01:32:49.740 01:32:55.490 Awaish Kumar: in Gemini like to get the data from all these sources. Is there any tool? Or we are going to

597 01:32:55.810 01:33:01.550 Awaish Kumar: do some like, write some of our own custom scripts get slack email, zoom.

598 01:33:01.550 01:33:05.020 Uttam Kumaran: We have, we have. We have custom scripts. So

599 01:33:06.870 01:33:15.453 Uttam Kumaran: let me so I’m gonna work on the proof of concept. And then I’m gonna this will be kind of on our platform team to work on. But let me

600 01:33:20.165 01:33:21.290 Uttam Kumaran: if you go

601 01:33:21.290 01:33:23.270 Awaish Kumar: That would be a great tool, like in itself.

602 01:33:23.880 01:33:24.660 Uttam Kumaran: Sorry.

603 01:33:25.650 01:33:28.768 Awaish Kumar: Yeah, if we are writing our own like scripts. And

604 01:33:29.290 01:33:32.910 Awaish Kumar: if like, build a framework it it. I’m saying that that

605 01:33:33.210 01:33:36.320 Awaish Kumar: that itself is going to be a great piece of tool

606 01:33:36.720 01:33:42.115 Uttam Kumaran: Oh, oh, my God, I mean, you know, like dude. This will solve so many of our fucking problems.

607 01:33:43.010 01:33:46.550 Uttam Kumaran: and guess what? It’s only possible because of AI right like.

608 01:33:46.780 01:33:53.740 Uttam Kumaran: So I so I in in slack. Yesterday I asked Casey who’s on our AI team. He wrote a script about

609 01:33:54.100 01:33:57.100 Uttam Kumaran: how to get all the messages from a slack channel.

610 01:33:57.490 01:34:00.145 Uttam Kumaran: So I’m gonna use this and

611 01:34:01.870 01:34:08.290 Uttam Kumaran: I’m gonna use this to basically run it locally and then break this into a project that we could all work on

612 01:34:10.230 01:34:10.690 Luke Daque: Not in this.

613 01:34:10.690 01:34:13.300 Uttam Kumaran: I think by Friday I have something that I can demo

614 01:34:14.920 01:34:15.969 Luke Daque: Yeah, cool.

615 01:34:16.830 01:34:22.660 Luke Daque: And then this could be a product. Right? Like, we can sell this to our customers like this. AI tool

616 01:34:22.990 01:34:28.540 Uttam Kumaran: Well, my hope is that we can make this anytime we interact with a customer. Our agent is there.

617 01:34:30.990 01:34:32.260 Uttam Kumaran: See what I mean?

618 01:34:32.680 01:34:35.739 Uttam Kumaran: The customer can ask the question to the agent first, st

619 01:34:37.610 01:34:38.460 Luke Daque: Right.

620 01:34:42.370 01:34:42.769 Uttam Kumaran: What do you

621 01:34:42.770 01:34:43.320 Luke Daque: But I guess

622 01:34:43.320 01:34:46.380 Uttam Kumaran: Like, what do you guys think about this section?

623 01:34:47.990 01:34:51.270 Uttam Kumaran: Cause this is the part where it’s like, this is all

624 01:34:51.270 01:34:53.050 Luke Daque: I’m still like trying

625 01:34:53.050 01:34:54.049 Awaish Kumar: Like the last

626 01:34:54.610 01:34:57.040 Uttam Kumaran: Yeah, the last part or, yeah, the last part

627 01:34:58.300 01:35:01.950 Awaish Kumar: Yeah, last part is like, that’s what like.

628 01:35:02.610 01:35:06.729 Awaish Kumar: we’ll answer the most of the question. And I think I really like that part

629 01:35:07.380 01:35:09.200 Uttam Kumaran: Well, cause like, I just don’t like

630 01:35:13.220 01:35:14.949 Uttam Kumaran: you guys can still see this right?

631 01:35:16.640 01:35:20.660 Luke Daque: We can see the implementation overview. Yeah, notion, right?

632 01:35:20.660 01:35:22.980 Uttam Kumaran: Oh, sorry. Okay. Let me share one. Hold on

633 01:35:29.740 01:35:36.470 Luke Daque: I’m still trying to wrap my head around around it like

634 01:35:36.470 01:35:38.060 Uttam Kumaran: Don’t worry about the AI part.

635 01:35:38.830 01:35:43.670 Luke Daque: I mean, yeah, like the the documentation, like itself.

636 01:35:43.890 01:35:46.130 Uttam Kumaran: Yeah, yeah. So I guess my question is like.

637 01:35:46.930 01:35:50.369 Uttam Kumaran: can we cut all? See how we have all these variations?

638 01:35:51.080 01:35:51.490 Luke Daque: Right.

639 01:35:51.490 01:35:56.540 Uttam Kumaran: I want to be able to cover all of this in one notion, Doc.

640 01:35:58.130 01:36:00.329 Uttam Kumaran: I don’t want to have this in spreadsheets

641 01:36:06.000 01:36:10.729 Uttam Kumaran: So, for example, maybe it’s 1 section per dashboard and Faqs per dashboard

642 01:36:11.560 01:36:18.210 Awaish Kumar: I think we need. This catalog might not like. Maybe Spreadsheet might not be the great tool, for that

643 01:36:18.340 01:36:25.320 Awaish Kumar: we might find some tool which can directly connect with the warehouse itself and create a catalog

644 01:36:25.910 01:36:30.630 Awaish Kumar: which can be easily shared with analyst team, so they can see what kind of tables

645 01:36:30.890 01:36:36.130 Awaish Kumar: and like fields are there. So we don’t have to create this sheet manually

646 01:36:36.130 01:36:38.919 Uttam Kumaran: But can I tell you one? Can I tell you? One problem I have with this

647 01:36:39.050 01:36:44.359 Uttam Kumaran: is our is like Armin. You gonna write this no dude, this is impossible.

648 01:36:46.010 01:36:51.450 Uttam Kumaran: So how are we gonna get the descriptions for everything, and stuff like that like this is where I don’t think

649 01:36:52.750 01:36:53.610 Uttam Kumaran: I don’t know

650 01:36:53.610 01:36:54.350 Luke Daque: Can we

651 01:36:54.350 01:36:54.799 Uttam Kumaran: I feel like

652 01:36:54.800 01:36:56.350 Luke Daque: Maybe utilize.

653 01:36:56.950 01:37:00.819 Luke Daque: Can we maybe utilize dbt, docs?

654 01:37:01.510 01:37:02.810 Luke Daque: But I guess

655 01:37:02.810 01:37:07.500 Uttam Kumaran: I’m still like I’m still confused. That like. Why, people can’t see this

656 01:37:07.630 01:37:09.179 Uttam Kumaran: and be like I get it

657 01:37:11.160 01:37:11.480 Luke Daque: Yes.

658 01:37:11.480 01:37:13.010 Uttam Kumaran: Confusing about this.

659 01:37:14.610 01:37:26.189 Uttam Kumaran: right? Like I get that. So first, st can I give you an example like if there’s a if there’s like a complicated piece of logic like, How how do you define profit? That’s gonna go in the FAQ for shopify?

660 01:37:26.680 01:37:32.979 Uttam Kumaran: But asking like, Do we really need every single column documented

661 01:37:35.320 01:37:38.069 Uttam Kumaran: cause this, this data is in snowflake.

662 01:37:39.210 01:37:42.179 Uttam Kumaran: And this data is in this data

663 01:37:42.180 01:37:43.110 Luke Daque: It’s redundant

664 01:37:43.110 01:37:46.939 Uttam Kumaran: Is it? Dbt is in Github. And this data is in

665 01:37:47.160 01:37:51.549 Uttam Kumaran: yeah, this all dbt, so like, why have? I’m fundamentally and I don’t.

666 01:37:51.720 01:37:56.439 Uttam Kumaran: When I saw this, I was like, there’s no way we’re gonna be able to maintain this

667 01:37:57.330 01:38:03.270 Luke Daque: Yeah, it’s very manual. And it’s yeah, like you mentioned, it’s redundant. Just it’s already in Snowflake. Why do we?

668 01:38:03.640 01:38:06.930 Luke Daque: We’ll need it there right?

669 01:38:07.760 01:38:08.470 Luke Daque: Like

670 01:38:08.470 01:38:09.150 Awaish Kumar: All right.

671 01:38:09.490 01:38:14.219 Uttam Kumaran: I could, I could do, I could. I could have the AI take the information. Schema put it in here.

672 01:38:14.510 01:38:16.059 Uttam Kumaran: But still, it’s like.

673 01:38:17.360 01:38:23.680 Uttam Kumaran: I think 20% of the columns need documentation. It’s just which 20% right like.

674 01:38:24.130 01:38:28.570 Uttam Kumaran: do we really, wanna we, the 3 of us, we want to spend time documenting these.

675 01:38:30.860 01:38:35.240 Uttam Kumaran: No way, you know.

676 01:38:38.700 01:38:43.320 Luke Daque: Yeah, I mean.

677 01:38:47.270 01:38:52.230 Luke Daque: yeah, it’s very redundant. The only difference here is that it has the

678 01:38:52.620 01:38:57.980 Luke Daque: logic, which is also even in like

679 01:38:58.170 01:39:05.000 Luke Daque: SQL. Syntax. So I’m not even sure if somebody who doesn’t have

680 01:39:05.260 01:39:10.690 Luke Daque: sequel knowledge would understand if it’s quite complicated.

681 01:39:11.260 01:39:15.330 Luke Daque: Right so.

682 01:39:17.690 01:39:22.749 Luke Daque: But we should have like Dbt. Has its documentation like it can be.

683 01:39:23.000 01:39:24.560 Luke Daque: We can document the

684 01:39:24.560 01:39:28.409 Uttam Kumaran: Dbt Docs dude the Dbt docs. All it does is this, it’s just this right

685 01:39:28.800 01:39:29.630 Luke Daque: It? Yeah.

686 01:39:29.630 01:39:32.570 Uttam Kumaran: We still have to fill it out. I guess what I’m saying is like

687 01:39:33.460 01:39:38.059 Uttam Kumaran: 20% of the things like need documentation.

688 01:39:38.210 01:39:40.419 Uttam Kumaran: Like, if I was to look at this table

689 01:39:40.540 01:39:43.000 Awaish Kumar: Probably we only need like

690 01:39:43.140 01:39:50.599 Uttam Kumaran: This, but fundamentally, what’s the FAQ. Here? The FAQ. Is, where can I go? See all the messages from an agent

691 01:39:50.830 01:39:53.529 Uttam Kumaran: go to fact? Messages use this column

692 01:39:55.700 01:39:58.179 Luke Daque: Yeah, that. And that’s

693 01:39:58.180 01:39:58.950 Uttam Kumaran: That’s FAQ,

694 01:39:58.950 01:39:59.730 Luke Daque: And that’s more

695 01:39:59.730 01:40:01.469 Uttam Kumaran: We still don’t need. We don’t need this

696 01:40:01.910 01:40:02.630 Luke Daque: Yeah.

697 01:40:05.720 01:40:14.100 Luke Daque: And that’s what the data analyst need. Even more like where to find the data as opposed to like how it’s being calculated

698 01:40:14.540 01:40:20.150 Uttam Kumaran: But my, my point is that I, if they’re like, Yeah, they but they can go to Github and find that anyways.

699 01:40:20.910 01:40:28.910 Uttam Kumaran: so once they find this, they can probably go to Github. But also, again, if we use AI to do the documentation, the AI should be able to

700 01:40:29.080 01:40:31.559 Uttam Kumaran: have access to the Github and know.

701 01:40:31.830 01:40:36.159 Uttam Kumaran: have access to this information schema about this table, and then and then write it

702 01:40:40.780 01:40:41.870 Awaish Kumar: And that’s true.

703 01:40:42.020 01:40:42.890 Uttam Kumaran: Right.

704 01:40:42.890 01:40:43.600 Luke Daque: Yeah.

705 01:40:43.770 01:40:47.750 Uttam Kumaran: Whether it’s whether it’s possible or not, like I’ll figure it out, bye.

706 01:40:52.010 01:40:53.449 Awaish Kumar: Me share with me.

707 01:40:53.730 01:41:01.690 Awaish Kumar: Yeah, like, that’s that we have to like push. For example, although, like we have.

708 01:41:02.600 01:41:07.900 Awaish Kumar: we have, like analyst team, have the extra snowflake or bigquery.

709 01:41:08.836 01:41:14.893 Awaish Kumar: They want to see some sheet where all the tables are listed. So it’s it’s also like,

710 01:41:15.420 01:41:21.210 Awaish Kumar: matter of pushing things like, okay, you can go in bigquery and see this.

711 01:41:25.560 01:41:27.810 Awaish Kumar: like the list of tables in in this

712 01:41:28.140 01:41:32.270 Awaish Kumar: productivity marks and everything in there. You can use anything

713 01:41:43.900 01:41:44.830 Luke Daque: Yeah.

714 01:41:56.670 01:42:01.360 Luke Daque: yeah, we can. We can can try that documentation

715 01:42:04.070 01:42:09.570 Luke Daque: and see where it gets us. Gets us like, FAQ style.

716 01:42:15.820 01:42:16.430 Uttam Kumaran: Okay.

717 01:42:19.360 01:42:25.000 Awaish Kumar: Yeah, like one of the thing which which big bigger enterprises

718 01:42:25.220 01:42:30.030 Awaish Kumar: use the catalogs for is that to volume, the schema changes.

719 01:42:30.250 01:42:38.380 Awaish Kumar: For example, if a table schema of a table changed for like 10 times, and we we

720 01:42:39.300 01:42:45.190 Awaish Kumar: deleted some columns which was being used previously, and then it’s easy to

721 01:42:45.400 01:42:49.859 Awaish Kumar: find out if we already have something for this table, we can bring it again.

722 01:42:49.960 01:42:51.809 Awaish Kumar: So this was the use case, like

723 01:42:52.250 01:42:52.630 Uttam Kumaran: Yeah.

724 01:42:52.630 01:42:56.559 Awaish Kumar: And then the bigger company. But I don’t know if it is needed for us

725 01:42:58.170 01:42:59.300 Luke Daque: Do we?

726 01:42:59.510 01:43:03.540 Luke Daque: Do we need a data dictionary to move

727 01:43:04.510 01:43:05.120 Awaish Kumar: Yeah, without?

728 01:43:05.290 01:43:07.240 Luke Daque: Be helpful as well

729 01:43:11.510 01:43:16.020 Awaish Kumar: Well, but that’s like similar thing like this sheet like it will have

730 01:43:16.200 01:43:19.030 Awaish Kumar: these fields and all these, like the

731 01:43:19.560 01:43:25.810 Uttam Kumaran: Yeah, no, that’s why that’s why I’m like, fundamentally, don’t don’t think about what the solution has been.

732 01:43:26.180 01:43:28.570 Uttam Kumaran: Think about the problem we’re solving.

733 01:43:29.120 01:43:33.660 Uttam Kumaran: We’re just need the answer. It doesn’t matter the format because we have AI

734 01:43:33.660 01:43:34.250 Luke Daque: Right.

735 01:43:35.090 01:43:38.990 Uttam Kumaran: The format is gonna be through the through slack agents.

736 01:43:43.140 01:43:53.059 Uttam Kumaran: All we needed to do is tell them, tell the answer. But I for me. I’m thinking about. What does the AI need to have? And I think one long document with everything will sort of solve it. For now

737 01:43:56.180 01:44:00.870 Awaish Kumar: Yes. But the major thing which I’m thinking is that the 1st question which comes from

738 01:44:01.020 01:44:04.049 Awaish Kumar: anyone is like for from analyst, is that

739 01:44:04.510 01:44:07.570 Awaish Kumar: okay, what fields we have for Zendesk?

740 01:44:08.070 01:44:15.739 Awaish Kumar: Right? Our documentation, FAQ, documentation or agent can answer the questions. But you cannot share, like

741 01:44:17.990 01:44:23.170 Awaish Kumar: everything upfront without without a question.

742 01:44:29.640 01:44:34.709 Uttam Kumaran: But I don’t know, Dude, what fields we have for Zendesk, but then, if it has access to all of our slacks.

743 01:44:35.860 01:44:36.930 Awaish Kumar: Okay, maybe

744 01:44:36.930 01:44:39.140 Uttam Kumaran: And it knows what. Zen no, no.

745 01:44:41.250 01:44:42.800 Uttam Kumaran: I mean, this is the one I’m gonna test

746 01:44:43.610 01:44:54.320 Awaish Kumar: Okay, yeah, yeah, yeah, like, okay, if if we can make agent, that intelligent, them is good

747 01:44:55.120 01:44:55.660 Uttam Kumaran: I mean, yeah.

748 01:44:55.660 01:44:58.990 Luke Daque: Yeah, if you can get context to our snowflake as well, and

749 01:44:59.180 01:45:01.939 Luke Daque: then you should be able to answer that question right?

750 01:45:03.100 01:45:07.379 Awaish Kumar: Yes, yeah. But yeah. Snowflake the schema.

751 01:45:07.840 01:45:12.689 Luke Daque: Or even yeah, or even Github

752 01:45:17.950 01:45:19.130 Awaish Kumar: Okay. Yeah.

753 01:45:26.930 01:45:28.459 Uttam Kumaran: Okay, perfect.

754 01:45:32.010 01:45:34.380 Uttam Kumaran: Alright, I’m gonna work on this a bit. Today.

755 01:45:34.610 01:45:36.590 Uttam Kumaran: If you have any ideas, let me know.

756 01:45:36.850 01:45:39.689 Uttam Kumaran: I’ll I’m gonna I’ll record this meeting, and sort of

757 01:45:40.170 01:45:43.139 Uttam Kumaran: send it to the team as well for feedback

758 01:45:45.510 01:45:46.910 Awaish Kumar: That’s good. Thank you.

759 01:45:47.150 01:45:48.959 Uttam Kumaran: But keep thinking about this like.

760 01:45:49.700 01:45:55.790 Uttam Kumaran: think, think, if you have a second, and think about what sort of abstractions we could do for our documentation.

761 01:45:55.930 01:46:03.519 Uttam Kumaran: I think, is like getting towards what it could be. One document that we can throw into AI, and it has everything right

762 01:46:04.090 01:46:04.660 Luke Daque: Yeah.

763 01:46:05.320 01:46:09.990 Uttam Kumaran: And I think Faqs is maybe the best sort of like

764 01:46:10.370 01:46:13.320 Uttam Kumaran: language medium for that that I know

765 01:46:15.600 01:46:16.719 Luke Daque: That makes sense.

766 01:46:17.900 01:46:19.929 Uttam Kumaran: To think about it. Okay, cool.

767 01:46:20.430 01:46:24.789 Uttam Kumaran: And then, yeah, always. If you have a second to do that shopify fees thing, let me know

768 01:46:28.690 01:46:31.479 Awaish Kumar: Cool to add that field right

769 01:46:31.480 01:46:32.110 Uttam Kumaran: Yeah.

770 01:46:32.870 01:46:36.649 Awaish Kumar: I I will add it, but I’m just going for dinner after that

771 01:46:36.650 01:46:38.080 Uttam Kumaran: Okay. Okay. No. Problem.

772 01:46:38.340 01:46:41.000 Awaish Kumar: Yeah. Thanks. Okay. Alright. Thanks. Guys.

773 01:46:41.670 01:46:43.230 Luke Daque: Sounds good. Thanks. See you