Meeting Title: Engineering Leads Sync Date: 2025-08-21 Meeting participants: Awaish Kumar, Samuel Roberts, Uttam Kumaran


WEBVTT

1 00:05:08.910 00:05:10.000 Samuel Roberts: A….

2 00:05:14.860 00:05:15.530 Awaish Kumar: Hello.

3 00:05:16.680 00:05:20.919 Samuel Roberts: Hey, sorry, we’re running a little late, we’re just finishing up the, ….

4 00:05:22.240 00:05:23.039 Uttam Kumaran: Hey, guys.

5 00:05:24.150 00:05:28.180 Samuel Roberts: Sorry, I didn’t get a chance to take a look at the doc, but maybe we can take….

6 00:05:28.180 00:05:30.380 Uttam Kumaran: A few minutes to add notes before we start.

7 00:05:30.380 00:05:34.049 Samuel Roberts: Yeah, that’d be good, I didn’t get a chance either yet.

8 00:05:49.400 00:05:51.589 Samuel Roberts: Of course not she isn’t playing nice.

9 00:05:52.070 00:05:53.000 Samuel Roberts: Here we go.

10 00:11:22.500 00:11:24.750 Uttam Kumaran: Okay, I think I’m ready.

11 00:11:26.080 00:11:26.790 Samuel Roberts: Boop.

12 00:11:31.730 00:11:32.860 Awaish Kumar: Promotes them.

13 00:12:15.910 00:12:17.190 Samuel Roberts: Excuse me.

14 00:12:27.760 00:12:29.820 Awaish Kumar: Should we start?

15 00:12:32.220 00:12:36.900 Uttam Kumaran: Yes, … Cool, so, yeah, I guess, like.

16 00:12:37.050 00:12:40.540 Uttam Kumaran: You think we should just stop the coaching stuff for Annie?

17 00:12:45.650 00:12:48.000 Awaish Kumar: So we have leadership.

18 00:12:49.060 00:12:51.230 Uttam Kumaran: Wait, can you say it again? I think it just cut out.

19 00:13:02.130 00:13:06.259 Uttam Kumaran: I guess more of my question was, like, yeah, we should just stop, right?

20 00:13:07.310 00:13:12.050 Awaish Kumar: Yeah, yeah, I think so, because… Like, she’s not…

21 00:13:13.080 00:13:16.509 Awaish Kumar: like, he’s more of an AE, and I…

22 00:13:16.920 00:13:20.049 Awaish Kumar: Don’t think she seems to be taking interest, because…

23 00:13:20.210 00:13:23.210 Awaish Kumar: kind of, I created a template, I wanted her

24 00:13:23.900 00:13:28.159 Awaish Kumar: To fill in, like, what she wanted to learn or discuss.

25 00:13:28.300 00:13:31.530 Awaish Kumar: In the next session, and it was not filled.

26 00:13:31.730 00:13:36.090 Awaish Kumar: So… And Robert already have started his, like.

27 00:13:36.540 00:13:41.300 Awaish Kumar: Scheduling sessions with her, so… I think we can stop.

28 00:13:41.490 00:13:43.200 Awaish Kumar: sessions with the Brian.

29 00:13:43.570 00:13:45.739 Awaish Kumar: And just keep it, with Robert.

30 00:13:48.300 00:13:48.990 Uttam Kumaran: Okay.

31 00:13:58.930 00:14:01.249 Awaish Kumar: Do you wanna go, Sam, on….

32 00:14:01.960 00:14:06.210 Samuel Roberts: Yeah, I mean, I thought you were telling me you put that in about Mustaf by KC.

33 00:14:06.710 00:14:07.680 Samuel Roberts: Bono.

34 00:14:08.220 00:14:10.260 Samuel Roberts: Engineering and systems thinking?

35 00:14:10.960 00:14:12.170 Samuel Roberts: I… Yeah.

36 00:14:12.280 00:14:12.890 Samuel Roberts: Yeah, go ahead.

37 00:14:12.890 00:14:13.550 Uttam Kumaran: Yeah, go.

38 00:14:13.750 00:14:14.429 Uttam Kumaran: But, ….

39 00:14:14.910 00:14:20.690 Samuel Roberts: Oh, I was gonna say, like, I think, yeah, I mean, I definitely see, you know, now that I know them a little bit better, I can see the, like.

40 00:14:20.830 00:14:27.030 Samuel Roberts: a little bit of the junior nature, but I… they definitely, like, seem like they get it. I… I’m wondering…

41 00:14:27.540 00:14:32.940 Samuel Roberts: what… What the best course of action is to, like, get them.

42 00:14:33.490 00:14:36.950 Uttam Kumaran: Yeah, like, a good… I guess a good way of thinking is, like.

43 00:14:37.160 00:14:42.850 Uttam Kumaran: how do I… how can we start to move them into higher leverage positions, and then….

44 00:14:42.850 00:14:43.330 Samuel Roberts: Yeah.

45 00:14:43.330 00:14:56.429 Uttam Kumaran: you know, like, basically fill folks under that, right? So, if… how can they start to move, not only in just, like, taking tickets and executing, but starting to plan technical designs.

46 00:14:56.530 00:15:08.979 Uttam Kumaran: starting to, like, lead development, you know, just, like, it doesn’t necessarily have to be, like, management, but it’s more of, like, being a partner to the PM versus.

47 00:15:08.980 00:15:09.320 Samuel Roberts: Yeah.

48 00:15:09.320 00:15:12.999 Uttam Kumaran: being an engineer on the task, right? It’s how I want to see their evolution.

49 00:15:15.190 00:15:18.429 Samuel Roberts: Yeah, I think that makes sense. You know, I…

50 00:15:18.860 00:15:34.610 Samuel Roberts: I have a sense that there may be ways to open that up as, like, you know, if a new, like, new project comes along, like, keeping them in the, like, bringing them into that process a little bit, where maybe it’s not like a, oh, here, just, you know, go throw together something, but it’s like a, let’s, let’s pair on that.

51 00:15:34.740 00:15:42.330 Samuel Roberts: You know, getting them thinking that way, asking them those… asking them the questions that they might not be asking themselves.

52 00:15:42.460 00:15:45.210 Samuel Roberts: To get them, like, comfortable with that.

53 00:15:45.400 00:15:57.529 Samuel Roberts: Because it is, it is definitely, like, I think they’re… they’re used to just, like, grabbing the tickets, but even… even then, I’m seeing them, like, thinking things through in ways that I’m like, okay, they’re… they’re capable of it, you know, I see it. ….

54 00:15:57.710 00:15:58.370 Uttam Kumaran: Definitely.

55 00:15:58.610 00:16:03.610 Samuel Roberts: But it’s a little hard for me right now, just because, like, some of these projects are just, like, ongoing, and I have a hard time, like, picturing, like.

56 00:16:04.370 00:16:05.629 Samuel Roberts: Where to jump into that?

57 00:16:05.630 00:16:13.170 Uttam Kumaran: Definitely, and certainly it would be on, like, some of the new projects, but also, again, like, some of the internal work, like, I want to see them leading, you know?

58 00:16:13.170 00:16:15.829 Samuel Roberts: Sure, okay, that’s actually, yeah, yeah, as we probably.

59 00:16:15.830 00:16:25.149 Uttam Kumaran: We’re the client there, so they can go meet with stakeholders, they can build the project plans, they can get our approval, they can execute, like… That’s a good point. Yeah.

60 00:16:26.470 00:16:29.319 Samuel Roberts: That’s a good point, I wasn’t thinking that way, but yeah, I mean…

61 00:16:30.200 00:16:33.919 Samuel Roberts: That’s probably the best way to do it to start, actually, now that you say it.

62 00:16:34.990 00:16:42.240 Samuel Roberts: It’s pretty much within our control, completely. It’s all, yeah, all internal that way. All right, yeah, I…

63 00:16:43.870 00:16:47.309 Samuel Roberts: I’ll give that some thought, in terms of, like, maybe I can start

64 00:16:47.640 00:16:54.049 Samuel Roberts: Hunting things to them a little bit, or pairing with them on some planning or something for more internal stuff when we have the, kind of.

65 00:16:54.740 00:17:02.320 Samuel Roberts: I mean, it also depends, like, how much breathing room we have for that sort of stuff, depending on how much, you know, stuff there is, just, like, throughput-wise client.

66 00:17:02.960 00:17:03.580 Samuel Roberts: ….

67 00:17:03.580 00:17:09.120 Uttam Kumaran: It’s just that where this would help us is just avoids needing a project manager, you know, for certain things.

68 00:17:09.569 00:17:11.569 Samuel Roberts: Right. Yeah, yeah.

69 00:17:14.010 00:17:14.790 Uttam Kumaran: Yeah.

70 00:17:15.109 00:17:17.470 Uttam Kumaran: Like, the faster that they can…

71 00:17:17.630 00:17:25.909 Uttam Kumaran: Also, start to see the business context, but also, like, learning a bit about, like, how to… how to meet with stakeholders, like, how to conduct demos.

72 00:17:26.520 00:17:31.130 Uttam Kumaran: like, leveling up communication a bit, you know? So, I think that’s something probably…

73 00:17:31.540 00:17:36.279 Uttam Kumaran: Sam, we can maybe think about a plan for them, and see how we can start to

74 00:17:36.430 00:17:38.750 Uttam Kumaran: You know, get them to that higher level, because…

75 00:17:39.380 00:17:45.730 Uttam Kumaran: I mean, they can continue to get… or the goal is to, like, have them go deeper on the technical side, right? So….

76 00:17:45.730 00:17:49.510 Samuel Roberts: That was… that was gonna be my next thing, was like, is this, you know, which way do they want to go with it, is the other.

77 00:17:49.510 00:17:52.740 Uttam Kumaran: It’s also… it’s also a good question for them. Yeah.

78 00:17:52.900 00:18:01.409 Uttam Kumaran: I think I’ve asked them in the past, but I would love to know, like, what direction they want to head, and how we can enable them to go do that.

79 00:18:01.680 00:18:02.730 Uttam Kumaran: Yeah.

80 00:18:03.550 00:18:04.520 Uttam Kumaran: you know.

81 00:18:06.430 00:18:07.050 Samuel Roberts: Okay.

82 00:18:08.850 00:18:10.249 Samuel Roberts: Yeah, I mean, I can…

83 00:18:10.420 00:18:19.849 Samuel Roberts: chat with them. I haven’t really gotten into that with them before, obviously, like, but I can do that a little bit and just get a sense of, like, where they’re, you know, …

84 00:18:20.100 00:18:24.719 Samuel Roberts: Yeah, let me… because I was thinking about even, like, setting up, like, one-on-ones, …

85 00:18:25.270 00:18:27.589 Samuel Roberts: In the next, like, week or two, just to, like.

86 00:18:27.790 00:18:36.959 Samuel Roberts: because I mean, I, you know, chat with them on the stand-up, but I would think it might be nice to just be, like, like, okay, let’s do a debrief, like, maybe not every week kind of thing, but at least, like, …

87 00:18:37.660 00:18:41.500 Samuel Roberts: Let me get a little bit of insight into this sort of side of them.

88 00:18:41.500 00:18:42.010 Uttam Kumaran: yet.

89 00:18:42.010 00:18:44.060 Samuel Roberts: Now that things are moving a little bit for me, so….

90 00:18:44.420 00:18:45.180 Uttam Kumaran: Yes.

91 00:18:46.330 00:18:46.890 Samuel Roberts: Cool.

92 00:18:47.860 00:18:50.659 Samuel Roberts: I think that’s good. And then, yeah, Ryan B, I don’t….

93 00:18:51.290 00:18:58.470 Uttam Kumaran: Yeah, Ryan is actually, like, very, very AI-driven. He’s on the marketing team. Okay.

94 00:18:58.960 00:19:10.980 Uttam Kumaran: I don’t know, I’m kind of trying to see, like, how we can enable him as well. Like, you know, he actually… he has a GPU, and he does a ton of stuff for how we’re… for our marketing side and our content.

95 00:19:12.370 00:19:21.859 Uttam Kumaran: I just feel like we should try to enable him and enable, you know, … His processes somehow, ….

96 00:19:21.860 00:19:22.750 Samuel Roberts: Yeah.

97 00:19:24.810 00:19:25.760 Samuel Roberts: Okay.

98 00:19:26.650 00:19:31.950 Samuel Roberts: is it worth to, like, have a chat with him and see, like… because I don’t know a ton about, like, what the marketing processes are like.

99 00:19:31.950 00:19:36.789 Uttam Kumaran: Yeah, I mean, I think it… I think if you could maybe grab, like, time with him just to chat.

100 00:19:37.010 00:19:43.970 Uttam Kumaran: Okay. And a little bit about how he’s using AI. He… out of all the teams, like, he’s… …

101 00:19:44.960 00:19:49.050 Uttam Kumaran: He’s the most AI-enabled, non-AI team person in the company, for sure.

102 00:19:49.550 00:19:50.220 Samuel Roberts: Okay.

103 00:19:53.290 00:19:53.670 Uttam Kumaran: Okay.

104 00:19:53.670 00:19:53.990 Samuel Roberts: Yeah.

105 00:19:53.990 00:19:54.600 Awaish Kumar: Okay, come on.

106 00:19:54.600 00:19:57.450 Samuel Roberts: do that. … But don’t.

107 00:19:59.690 00:20:00.759 Samuel Roberts: Oh, excuse me.

108 00:20:07.150 00:20:13.120 Awaish Kumar: Okay, on the excellence side, I… Wanted to have, …

109 00:20:13.900 00:20:19.880 Awaish Kumar: like, like, for example, for the demo sessions, we created an ocean talk, Which exactly…

110 00:20:20.140 00:20:23.660 Awaish Kumar: Defines, who’s… who will be the next, so…

111 00:20:24.190 00:20:28.760 Awaish Kumar: Here, we can kind of create similar Notion documents, and we define

112 00:20:29.460 00:20:32.140 Awaish Kumar: Honors for each of the tools.

113 00:20:32.820 00:20:38.819 Awaish Kumar: Metaplane, Dexter, Windmill, GitHub Actions, and then…

114 00:20:39.730 00:20:43.099 Awaish Kumar: The owner is responsible to, like, be a first

115 00:20:43.800 00:20:47.629 Awaish Kumar: A line of defense, like, the further threat, at least for dry age.

116 00:20:48.940 00:20:53.150 Awaish Kumar: For any errors which come from that tool, and then,

117 00:20:54.140 00:21:00.029 Awaish Kumar: Yeah, and then it goes from there, if it needs a longer ticket, or it should be resolved right away.

118 00:21:03.560 00:21:04.240 Uttam Kumaran: Okay.

119 00:21:04.950 00:21:12.330 Uttam Kumaran: Yeah, I do think we should have tool owners. … Maybe we can…

120 00:21:16.360 00:21:18.150 Uttam Kumaran: Yeah, I agree.

121 00:21:21.570 00:21:26.459 Uttam Kumaran: So… I mean, we could put it… we could have… we could create our own, …

122 00:21:30.090 00:21:33.539 Uttam Kumaran: Yeah, I think we should just create this in Notion, ….

123 00:21:35.060 00:21:37.470 Awaish Kumar: We can just create a notion and announce it.

124 00:21:37.610 00:21:38.720 Awaish Kumar: Everyone?

125 00:21:38.940 00:21:39.570 Uttam Kumaran: Okay.

126 00:21:39.700 00:21:44.799 Awaish Kumar: Like, who is the owner, and, what they are responsible for.

127 00:21:46.450 00:21:48.030 Awaish Kumar: Black and the… Yep.

128 00:21:48.940 00:21:50.500 Uttam Kumaran: Okay, okay, that would be great.

129 00:21:51.860 00:22:00.179 Samuel Roberts: Yeah, my only question there is, like, I’m still, like, familiarizing myself with, like, Dagster, and then I think, Lutam, you brought up the Dagster windmill, like, failure we’ve seen in the, …

130 00:22:00.760 00:22:02.119 Samuel Roberts: In Slack, at least.

131 00:22:02.200 00:22:06.760 Uttam Kumaran: I’m curious, like… It’s, like, simple Python jobs that need to get debugged, but, like.

132 00:22:06.760 00:22:07.780 Samuel Roberts: Yeah.

133 00:22:08.220 00:22:18.890 Uttam Kumaran: I think it’s helpful to know, like, who the owner is, and, like, who… who can, like… because those are platforms, right? Like, anyone on the team that wants to orchestrate can use Dagster to do that.

134 00:22:19.160 00:22:24.160 Uttam Kumaran: So it’s not necessarily, like, the owner is going to maybe triage, but… or the owner….

135 00:22:24.160 00:22:26.140 Samuel Roberts: That’s what I was getting at, yeah.

136 00:22:26.140 00:22:32.470 Uttam Kumaran: But the owner certainly can be like, hey, this is failing, I’m gonna turn it off, so whatever this is supporting needs to get fixed.

137 00:22:32.860 00:22:50.529 Samuel Roberts: Okay, yeah, that’s actually a good way to think about it, because I was… that’s what I was worried about, because I mean, like, I know, like, we’re using DAC for a few things here, like, on the, you know, embeddings and stuff, and I just, like, if the tool owner doesn’t realize what’s being used for what, where, if it’s more detailed, or, like, down the tree a little bit, from the owner.

138 00:22:53.640 00:22:59.779 Samuel Roberts: But they could be the one responsible for, like, keeping an eye on, like, oh, this is failing a bunch, you know, who did this or something, maybe.

139 00:22:59.780 00:23:00.830 Awaish Kumar: Yeah. I don’t know.

140 00:23:02.370 00:23:06.120 Awaish Kumar: If you have the owner, is it, like, you will be responsible for looking at the…

141 00:23:07.240 00:23:12.810 Awaish Kumar: Errors, and creating the tickets, maybe, and assigning it to the right person.

142 00:23:13.240 00:23:14.430 Samuel Roberts: That’s… yeah, I think that makes sense.

143 00:23:14.730 00:23:15.450 Awaish Kumar: Yep.

144 00:23:22.310 00:23:24.759 Awaish Kumar: Yeah, for, like, I saw, like.

145 00:23:24.890 00:23:30.270 Awaish Kumar: maybe we thought we added this data range and dashboarding, so I just had a…

146 00:23:30.580 00:23:34.110 Awaish Kumar: like, an introductory session with Andre.

147 00:23:34.430 00:23:36.849 Awaish Kumar: And, he looked, like, promising.

148 00:23:37.490 00:23:37.880 Uttam Kumaran: Okay.

149 00:23:37.880 00:23:38.540 Awaish Kumar: I’m fine.

150 00:23:39.560 00:23:44.500 Awaish Kumar: To me, so yeah, we can consider him. I, like, it was a 30-minute session, I couldn’t go…

151 00:23:45.490 00:23:51.809 Awaish Kumar: Deeper on any technical things, but yeah, like, just… just hearing the experience and,

152 00:23:52.280 00:23:56.389 Awaish Kumar: he seemed to have, like, very good experience with Tableau, Power BI.

153 00:23:57.010 00:23:59.239 Awaish Kumar: Great. Client Fishing Group, yeah.

154 00:24:00.360 00:24:01.520 Uttam Kumaran: Okay, okay, perfect.

155 00:24:02.030 00:24:09.060 Uttam Kumaran: Okay, that’s really, really great to hear. Yeah, I think we’re also gonna put up an additional job post for a senior AE.

156 00:24:09.620 00:24:11.989 Uttam Kumaran: Kind of exhausted my networks.

157 00:24:12.190 00:24:14.739 Uttam Kumaran: So, want to put that up. …

158 00:24:15.050 00:24:17.369 Uttam Kumaran: And then Voshdev said he can be part-time.

159 00:24:17.780 00:24:19.710 Uttam Kumaran: Perfect, so…

160 00:24:20.240 00:24:30.559 Uttam Kumaran: like, I don’t… like, if I can get out of client work, I can take on the DE stuff. It’s, like, so easy, but I just, like, I’m spending a lot of time on other stuff, so…

161 00:24:30.670 00:24:34.590 Uttam Kumaran: I’m sort of hopeful that, like, if we can get me out of urban stems.

162 00:24:36.070 00:24:37.939 Awaish Kumar: Yeah, day work is….

163 00:24:39.170 00:24:40.370 Uttam Kumaran: It’s not that much.

164 00:24:40.370 00:24:48.469 Awaish Kumar: I, I, I can… I’m looking into, like, eating escapes sometime, and, Casey…

165 00:24:48.870 00:24:54.959 Awaish Kumar: is also, like, looking into Slack and other stuff in the Dagstar, so… It can be handled.

166 00:24:55.660 00:24:56.300 Uttam Kumaran: Yeah.

167 00:24:57.310 00:24:58.020 Uttam Kumaran: …

168 00:24:58.520 00:25:09.860 Uttam Kumaran: And then, yeah, there’s… we were interviewing, one more AI engineer. He’s actually based here in Boston, but he’s gonna be moving here to Austin soon.

169 00:25:10.190 00:25:15.469 Uttam Kumaran: Right now, overall, our budget for people is really aligned towards bringing on

170 00:25:15.610 00:25:19.830 Uttam Kumaran: another data analyst hire, and another PM.

171 00:25:20.040 00:25:24.059 Uttam Kumaran: So we don’t really have more budget for, like, more engineering. Also, like.

172 00:25:24.060 00:25:24.700 Samuel Roberts: I think.

173 00:25:25.040 00:25:41.289 Uttam Kumaran: Sam, with you, Casey, and Mustafa, we’re kind of covered on, like, existing clients. So, kind of like… but I think he’s really, really great, so I’m kind of asking him to see, maybe he can just spend, like, a couple hours a week with us, and just, like, join, like.

174 00:25:41.460 00:25:46.669 Uttam Kumaran: sprint reviews or brainstorming sessions. He’s not, like, on a super time crunch.

175 00:25:46.830 00:25:51.280 Uttam Kumaran: I think he does want to work with us longer term, so I would like to find a way to kind of keep him in our…

176 00:25:51.550 00:25:53.010 Uttam Kumaran: In our orbit.

177 00:25:53.580 00:25:54.589 Samuel Roberts: Yeah, that’s smart.

178 00:25:55.900 00:26:06.549 Uttam Kumaran: But more of my… more of my priority is, like, getting you to full-time, and, like, getting a couple more clients before we can… and getting the… the sort of data side figured out, because I think on the AI side.

179 00:26:06.550 00:26:08.760 Samuel Roberts: Our team is really good, so….

180 00:26:08.800 00:26:10.379 Uttam Kumaran: It’s not that bad.

181 00:26:10.620 00:26:12.390 Uttam Kumaran: Not as much of a concern.

182 00:26:13.270 00:26:22.780 Samuel Roberts: Yeah, it’s definitely one of those things where, like, if you see, like, good talent, you don’t want to let it go completely, but if you’re not… Yeah. Yeah. So yeah, I mean, if there’s a way to, like, you know, loop him in a little bit and keep him kind of…

183 00:26:22.990 00:26:24.949 Samuel Roberts: On the string a little bit, and not…

184 00:26:25.420 00:26:29.860 Samuel Roberts: You know, commit yet, but bring on a few more clients and everything.

185 00:26:30.160 00:26:32.470 Samuel Roberts: I’m down for that, yeah, even if it’s just, like.

186 00:26:32.820 00:26:38.260 Samuel Roberts: Brainstorming stuff, but that actually might be a good way to, like, maybe we can commit some time to, like, internal stuff, even.

187 00:26:38.390 00:26:40.800 Samuel Roberts: And bring him in, like, where it’s not…

188 00:26:40.950 00:26:44.690 Samuel Roberts: Client works, necessarily, so it’s not, like, gonna be… You know.

189 00:26:45.080 00:26:47.290 Samuel Roberts: Time crunchy the same way.

190 00:26:48.380 00:26:52.950 Samuel Roberts: But getting some other input is probably something good we can do with some of the internal platform stuff.

191 00:26:54.350 00:26:58.080 Samuel Roberts: Yeah. If he’s down for a few hours a week, I’ll schedule around that.

192 00:27:02.920 00:27:03.930 Uttam Kumaran: Okay, great.

193 00:27:04.070 00:27:05.160 Uttam Kumaran: …

194 00:27:06.580 00:27:14.720 Uttam Kumaran: Yeah, and then on the AI side, so, yesterday I was doing some work on, like, just looking at AI PR reviews, so I got a cursor.

195 00:27:15.100 00:27:18.750 Uttam Kumaran: And using cursors like BugBot, to actually….

196 00:27:18.750 00:27:19.540 Samuel Roberts: Right.

197 00:27:19.540 00:27:29.809 Uttam Kumaran: edit PRs, and so you can invoke it, like, you can go into GitHub and, actually, like, maybe… oh, wait, I can just share so that three of us are…

198 00:27:31.490 00:27:34.610 Uttam Kumaran: are… And see this really quickly.

199 00:27:38.420 00:27:40.479 Awaish Kumar: I saw your… when I was your PR.

200 00:27:40.950 00:27:44.760 Uttam Kumaran: Cool, yeah, yeah, yeah, I was just testing yesterday, so…

201 00:27:45.230 00:27:48.599 Uttam Kumaran: And I think I’ll make… I think let me just make sure that you guys are…

202 00:27:51.900 00:27:53.570 Uttam Kumaran: admins.

203 00:28:02.500 00:28:09.930 Uttam Kumaran: Cool, so in BugBot, it’s here, I’ve enabled it for just the DAX or pipelines.

204 00:28:10.070 00:28:11.220 Uttam Kumaran: repo.

205 00:28:11.550 00:28:17.150 Uttam Kumaran: So it’ll run on all PRs, … And…

206 00:28:17.370 00:28:21.290 Uttam Kumaran: you can also invoke it, like, you can just say at BugBot or at cursor.

207 00:28:21.290 00:28:21.800 Samuel Roberts: Oh, cool.

208 00:28:21.800 00:28:24.530 Uttam Kumaran: And it will… it will,

209 00:28:25.970 00:28:28.619 Uttam Kumaran: Sort of give a… give a review or give comments.

210 00:28:32.950 00:28:36.180 Uttam Kumaran: So, I feel like it’s a little bit in a step in the right direction. I could.

211 00:28:36.180 00:28:36.610 Samuel Roberts: I was figuring out.

212 00:28:36.780 00:28:42.560 Uttam Kumaran: I think great for, like, writing the PR description, ….

213 00:28:43.420 00:28:49.359 Uttam Kumaran: You know, but… That’s fine I think there’s probably something we can build

214 00:28:49.610 00:28:57.440 Uttam Kumaran: relating linear to that, if we need to. Main thing is I just want, like, one AI review step on all PRs.

215 00:28:57.960 00:28:58.640 Samuel Roberts: Yeah.

216 00:28:58.640 00:29:09.090 Uttam Kumaran: It will save us a little bit of time. The other thing is, like, yeah, I think Demolade was right. On the data side, we just need a lot more CICD, like, we just need to be able to test

217 00:29:09.760 00:29:12.150 Uttam Kumaran: To test things in staging.

218 00:29:13.110 00:29:16.599 Uttam Kumaran: And see impact, so… so something that we’re gonna have to…

219 00:29:17.800 00:29:21.450 Uttam Kumaran: figure out, but I think this is a better… good step in, like.

220 00:29:21.590 00:29:25.120 Uttam Kumaran: in that direction of having AI do the first PR review for stuff.

221 00:29:27.240 00:29:28.760 Awaish Kumar: That’s, like, derive.

222 00:29:29.230 00:29:31.029 Awaish Kumar: Phew, you want to ask you?

223 00:29:32.550 00:29:33.360 Uttam Kumaran: Yeah.

224 00:29:36.220 00:29:40.770 Samuel Roberts: … Sorry, lost my train of thought. I had something about the PR.

225 00:29:41.610 00:29:50.170 Samuel Roberts: and linear… Oh, at some point, I had been using, some add-on in VS Code. This is, like, precursor to open…

226 00:29:50.370 00:29:56.440 Samuel Roberts: PRs into GitHub, So I’m wondering if there’s a way to, like… I mean…

227 00:29:56.630 00:30:01.219 Samuel Roberts: if we were living in Cursor and you can, like, pull in linear stuff from an MCP and get a PR written

228 00:30:02.090 00:30:06.229 Samuel Roberts: There… Okay, I’m gonna have to think about that a little bit more.

229 00:30:06.230 00:30:11.619 Uttam Kumaran: Yeah, no, I think you can go into Kirscher, do the linear MCP, bring the ticket in, execute.

230 00:30:12.440 00:30:15.280 Uttam Kumaran: And then I think Cursor can automatically write the PR.

231 00:30:15.900 00:30:20.469 Samuel Roberts: That’s what I’m wondering, because I know, I know there, like, you can… I, you know, I’m just not sure if it’ll…

232 00:30:21.590 00:30:25.620 Samuel Roberts: do the PO… yeah, I’m… that’s what I’m wondering. I don’t know if it, like, ties to…

233 00:30:25.870 00:30:30.329 Samuel Roberts: Because, like, the PRs are, like, on top of just a regular merge, so I’m not sure if it knows.

234 00:30:30.430 00:30:32.430 Samuel Roberts: Yeah…

235 00:30:36.000 00:30:40.189 Samuel Roberts: Bounce in advance, prompting my cursor up. Yeah, so if it’s CLI stuff, maybe, maybe there’s a way.

236 00:30:41.970 00:30:46.090 Samuel Roberts: But that’s at least, I mean, honestly, just even thinking that way, like, we might be able to get something that’s, like.

237 00:30:47.110 00:30:50.360 Samuel Roberts: if someone opens a PR, maybe we have something trigger a, you know.

238 00:30:50.360 00:30:51.720 Uttam Kumaran: Yes. Yeah.

239 00:30:51.720 00:30:56.490 Samuel Roberts: trigger something, and then it fills in the description for you, at least. You know what I mean? Just, like, as a first pass, like…

240 00:30:56.550 00:31:13.070 Samuel Roberts: get something that’s just, like, it might not be as automated as, like, it generates and everything, but at least, like, if you open a PR with no description, maybe it looks at it and compares to linear or something. That’s definitely worth thinking about, if nothing’s, like, out of the box for us.

241 00:31:13.260 00:31:21.100 Samuel Roberts: But that’s… okay, that’s cool, but yeah, bug… Interesting. Okay, cool, that’s… But run.

242 00:31:21.360 00:31:24.310 Uttam Kumaran: Cool. Yeah, the only other thing is, …

243 00:31:24.450 00:31:28.860 Uttam Kumaran: I just wanted to create a rule set so that maybe we can make sure that all

244 00:31:29.590 00:31:34.690 Uttam Kumaran: Branches have the same… have the, … Linear naming convention.

245 00:31:35.340 00:31:36.270 Samuel Roberts: Mmm.

246 00:31:37.220 00:31:39.420 Uttam Kumaran: That way, we can make sure it always gets linked.

247 00:31:41.820 00:31:42.910 Samuel Roberts: Yeah….

248 00:31:43.870 00:31:44.500 Uttam Kumaran: Yeah.

249 00:31:45.680 00:31:46.760 Samuel Roberts: That’s… okay.

250 00:31:48.380 00:31:50.390 Samuel Roberts: There’s gotta be a way to… yeah.

251 00:31:54.050 00:31:59.479 Samuel Roberts: Yeah, I can probably get that set up on… Something… I think you can do….

252 00:31:59.480 00:32:02.250 Uttam Kumaran: GitHub admin, like, we just have to….

253 00:32:02.800 00:32:05.529 Samuel Roberts: Enforce it, yeah. Like, they put the rule in.

254 00:32:06.650 00:32:07.160 Uttam Kumaran: Yeah.

255 00:32:07.160 00:32:13.040 Samuel Roberts: … Okay, I can take a look into that if you want.

256 00:32:13.490 00:32:16.769 Samuel Roberts: Is that, like, a per-repo basis, or is that an organizational basis?

257 00:32:16.770 00:32:18.490 Uttam Kumaran: It’s on a word basis.

258 00:32:18.660 00:32:19.490 Samuel Roberts: Okay.

259 00:32:23.780 00:32:25.780 Samuel Roberts: Let me follow… I’m not sure what I’m…

260 00:32:27.480 00:32:32.550 Samuel Roberts: Where’s, like, settings for the org? Maybe I don’t have access to the settings for the org, actually.

261 00:32:33.260 00:32:38.999 Samuel Roberts: That might not be… I don’t know what my role is in the org.

262 00:32:41.700 00:32:46.969 Samuel Roberts: automation team, yeah. Yeah, I’m gonna be… I might be able to do that.

263 00:32:47.890 00:32:49.240 Awaish Kumar: At this point.

264 00:32:50.050 00:32:54.910 Awaish Kumar: I added these, … As the next plan of action, ….

265 00:32:54.910 00:32:55.540 Samuel Roberts: Okay.

266 00:32:55.680 00:32:56.390 Samuel Roberts: Cool.

267 00:32:58.090 00:33:00.439 Awaish Kumar: Is that… am I missing something?

268 00:33:09.280 00:33:10.700 Samuel Roberts: In a process…

269 00:33:12.370 00:33:18.059 Samuel Roberts: I had something else, but… oh, I was gonna reach out to Ryan, but that’s just… I’m just gonna chat with him, so I wouldn’t worry about that.

270 00:33:19.890 00:33:21.200 Uttam Kumaran: …

271 00:33:24.230 00:33:28.940 Uttam Kumaran: Yeah, the other thing is, like, I think we’ve got to start building some runbooks for common tasks.

272 00:33:30.520 00:33:32.600 Samuel Roberts: Oh, yeah, sorry to put that there, yeah.

273 00:33:40.360 00:33:41.350 Uttam Kumaran: …

274 00:33:44.930 00:33:52.400 Uttam Kumaran: Yeah, I just think, like, we need to start improving, like, kind of just… Just overall documentation.

275 00:33:55.050 00:33:56.779 Uttam Kumaran: Think a little bit about how we do that.

276 00:33:59.080 00:34:04.880 Samuel Roberts: Like, are there specific areas, like, is it, like, specifically, like, the data stuff, or is it, like.

277 00:34:06.230 00:34:10.179 Samuel Roberts: Things, like, non-engineering related? Or, like, what are you thinking, like…

278 00:34:10.679 00:34:15.860 Samuel Roberts: Coming task-wise. Because, like, I think that’s probably helpful in a lot of ways, if… because, like.

279 00:34:16.060 00:34:25.659 Samuel Roberts: from the AI, like, eventual automation side, like, having that documented is gonna be big, because there’s a lot of things I don’t fully understand myself… yet.

280 00:34:25.800 00:34:27.380 Samuel Roberts: Company-wise, but….

281 00:34:27.389 00:34:28.099 Uttam Kumaran: Yeah, I agree.

282 00:34:28.100 00:34:29.209 Samuel Roberts: Where your head’s at with that.

283 00:34:29.210 00:34:36.959 Uttam Kumaran: I think on the AI side, it’s like creating new prompts, or creating new nodes, or things like that, and I think on the data side, there’s a lot, like…

284 00:34:37.370 00:34:46.350 Uttam Kumaran: updating column, like, we just need to have, like, SOPs for common tasks before we’re ever going to get to the point where we could use AI to automate it.

285 00:34:47.260 00:34:48.260 Samuel Roberts: Yeah.

286 00:34:50.800 00:34:52.100 Uttam Kumaran: You know, so….

287 00:34:57.800 00:34:58.790 Awaish Kumar: Okay.

288 00:35:01.240 00:35:03.889 Uttam Kumaran: Okay, cool, so…

289 00:35:08.960 00:35:13.149 Uttam Kumaran: I feel pretty good overall, I think, yeah, I… I wanna…

290 00:35:13.820 00:35:22.820 Uttam Kumaran: I want to continue to have Sam, you kind of get a little bit closer to Mustafa and Casey and see how we can continue to invest in them and level them up.

291 00:35:23.140 00:35:28.689 Uttam Kumaran: You know, ideally, like, I think they’re… they’re both smart enough to start to also, like.

292 00:35:28.900 00:35:35.090 Uttam Kumaran: really meet with clients and PM stuff. Like, I want them to become really, really great, holistic engineers.

293 00:35:35.410 00:35:41.749 Uttam Kumaran: But also understand, like, I think both of them have previously indicated, I just want to keep going deeper, technically, but…

294 00:35:41.860 00:35:43.960 Uttam Kumaran: I also want them to be able to, sort of.

295 00:35:44.340 00:35:50.090 Uttam Kumaran: work directly with clients, and get requirements and build, and… and improve that, you know, so….

296 00:35:50.090 00:35:51.010 Samuel Roberts: Yeah.

297 00:35:52.670 00:35:53.710 Uttam Kumaran: … Okay.

298 00:35:54.960 00:35:58.429 Awaish Kumar: Oh yeah, one more thing about… I have scheduled a meeting

299 00:35:58.610 00:36:01.400 Awaish Kumar: Farimo, with three of us.

300 00:36:03.570 00:36:04.540 Uttam Kumaran: Okay.

301 00:36:05.430 00:36:06.250 Awaish Kumar: Wow.

302 00:36:06.900 00:36:13.159 Awaish Kumar: Yeah, we can discuss, right? But they are just sharing ERD diagram, like, there’s….

303 00:36:14.450 00:36:15.080 Uttam Kumaran: Okay.

304 00:36:15.290 00:36:16.499 Awaish Kumar: There’s nothing else.

305 00:36:18.180 00:36:21.360 Awaish Kumar: It just tells us the, like, the data models, but yeah.

306 00:36:21.610 00:36:25.319 Awaish Kumar: It doesn’t tell us about the architecture of the system, so…

307 00:36:25.430 00:36:27.219 Awaish Kumar: At least we would need that.

308 00:36:27.570 00:36:32.729 Awaish Kumar: But that’s what, … Stuff will be helping with, like, improving the architecture?

309 00:36:34.990 00:36:35.590 Uttam Kumaran: Okay.

310 00:36:37.210 00:36:39.509 Awaish Kumar: I already, like, sent a message.

311 00:36:39.770 00:36:42.430 Awaish Kumar: Totally on about that, and Chris.

312 00:36:42.990 00:36:46.139 Awaish Kumar: In the channel, but yeah, if you want to bump… bump that up, yeah.

313 00:36:46.140 00:36:47.490 Uttam Kumaran: I’ll be bumping, yeah.

314 00:36:51.880 00:36:54.279 Uttam Kumaran: Okay, great. All right.

315 00:36:54.400 00:36:56.079 Uttam Kumaran: Thank you, guys, appreciate it.

316 00:36:56.830 00:36:57.210 Samuel Roberts: Yeah.

317 00:36:58.970 00:37:00.090 Awaish Kumar: Okay, thank you.