2026-04-15_brainforge_x_cta_project_planning_c4d9f6da

Meeting Title: Brainforge x CTA Project Planning Date: 2026-04-15 Meeting participants: Uttam Kumaran, Katherine Bayless

---

WEBVTT

1 00:00:31.880 ⇒ 00:00:32.810 Katherine Bayless: Hello!

2 00:00:33.640 ⇒ 00:00:34.670 Uttam Kumaran: Hi!

3 00:00:34.670 ⇒ 00:00:35.470 Katherine Bayless: How’s it going?

4 00:00:35.960 ⇒ 00:00:37.340 Uttam Kumaran: Good, how’s everything.

5 00:00:37.680 ⇒ 00:01:02.519 Katherine Bayless: quite good. In fact, well, also just total chaos all the time. But, no, I’m, I’m, I’m grinning ear to ear because I just got out of, talking to… so Christina’s team is, like, the leadership under finance and administration, so it’s, like, legal, finance, IT, although Jay’s out this week, and then the foundation guy, right? So it’s, like, you know, all the ops people, right? And so I was demoing Snow

6 00:01:02.520 ⇒ 00:01:08.150 Katherine Bayless: like, for them, so that they could finally get a chance to see it. And as I, like, walked through the stuff.

7 00:01:08.150 ⇒ 00:01:22.760 Katherine Bayless: the finance guy was, like, saw, like, a stat on the screen. He’s like, that can’t be right. And I was like, yeah, I don’t think it is. I was like, I think, actually, I’m guessing that Coco pulled that from the remembers for, like, essentially, like, copy of the transaction data, and I was like.

8 00:01:22.760 ⇒ 00:01:35.530 Katherine Bayless: I’ve seen recently that we’ve been piping things in as if they were transactions versus activities, and he’s like, which is the dumbest thing. I don’t… we only did that because the teams want the data, and that was the only way to do it, and like, and I was like.

9 00:01:36.040 ⇒ 00:01:38.680 Katherine Bayless: So have I got a solution for you, right?

10 00:01:38.680 ⇒ 00:01:39.320 Uttam Kumaran: Great.

11 00:01:39.320 ⇒ 00:01:44.219 Katherine Bayless: you… the light bulb went off so bright, I mean, I got a tan. It was amazing.

12 00:01:44.220 ⇒ 00:01:57.250 Uttam Kumaran: Wonderful. I’m just reflecting on, like, when we first started talking in December and, like, January. This was really good. I mean, it’s good progress for, like, right now, I think, like, still with, like, 4 months to go.

13 00:01:57.450 ⇒ 00:01:58.120 Katherine Bayless: Before…

14 00:01:58.120 ⇒ 00:02:00.350 Uttam Kumaran: Or stuff starts to ramp up, you know?

15 00:02:00.820 ⇒ 00:02:11.440 Katherine Bayless: Right? And then, like, legal, she’s like, I mentioned, like, Snowflake has, you know, native capability to sell data, and she’s like, how soon can we start doing this? And we’re like, honey, you’re legal, right? You know?

16 00:02:11.440 ⇒ 00:02:30.570 Uttam Kumaran: Yeah, that’s actually a great thing, too. I… we… I did it at another company, and actually, when I first started Brainforge, one of my ideas was, like, I wanted to get into data marketplace, and so I sold… I sold, like, $100 worth of data on there. I sold some SEC-related data sets, like, a few years ago.

17 00:02:30.570 ⇒ 00:02:31.079 Katherine Bayless: Yeah, yeah.

18 00:02:31.080 ⇒ 00:02:38.510 Uttam Kumaran: Yeah, we’re… we actually work a lot with some of, like, the companies, like, similar to some of our vendors that, like.

19 00:02:38.650 ⇒ 00:02:49.620 Uttam Kumaran: broker data through there, do snowflake sharing. We have another client where they actually… they’re like a… they’re like a tax firm, and they want to, like, bring in information, model it, and then send it back.

20 00:02:50.020 ⇒ 00:02:51.289 Katherine Bayless: That’s a.

21 00:02:51.290 ⇒ 00:02:52.780 Uttam Kumaran: application, basically.

22 00:02:53.030 ⇒ 00:02:57.620 Katherine Bayless: I think that would be probably where we start, because we currently do, I think it’s…

23 00:02:57.750 ⇒ 00:03:04.960 Katherine Bayless: semi-annual, or, you know, twice a year, whichever way that works, a forecast, and, like, we sell the report, but we would like.

24 00:03:04.960 ⇒ 00:03:05.350 Uttam Kumaran: Okay.

25 00:03:05.350 ⇒ 00:03:10.150 Katherine Bayless: the data as well, right? And it’s just, you know, crowdsourced, available data, right?

26 00:03:10.150 ⇒ 00:03:10.500 Uttam Kumaran: Yes.

27 00:03:10.500 ⇒ 00:03:13.550 Katherine Bayless: assemble it and sell it for, you know, $50,000, maybe?

28 00:03:13.550 ⇒ 00:03:22.840 Uttam Kumaran: Yes, yeah, I think so too, and you just deliver, and then also being able to deliver it that way gives their team, they’re like, oh, this is nice, they’re already on Snowflake.

29 00:03:22.840 ⇒ 00:03:33.430 Katherine Bayless: Right? Right? Right, exactly. So, yeah. Like, Christina asked me yesterday in our one-on-one, like, how I was feeling bandwidth-wise, and I was like, oh, we are 100% in the danger zone, but I…

30 00:03:34.480 ⇒ 00:03:35.390 Katherine Bayless: It’s okay.

31 00:03:35.390 ⇒ 00:03:41.960 Uttam Kumaran: I thought you were gonna say we were 100% good, and like, I’m… No, no, we’re…

32 00:03:41.960 ⇒ 00:03:47.490 Katherine Bayless: Full danger zone, but… I think it’s a good thing. Yeah.

33 00:03:47.490 ⇒ 00:03:52.319 Uttam Kumaran: I think… but I think it’s dangerous… I think what… however, I… I feel like the politics…

34 00:03:52.470 ⇒ 00:03:54.809 Uttam Kumaran: You’re starting to be able to clear that up.

35 00:03:55.030 ⇒ 00:03:55.870 Katherine Bayless: Yeah.

36 00:03:55.870 ⇒ 00:04:00.119 Uttam Kumaran: Versus it still being an issue in addition to, like, all the building.

37 00:04:00.310 ⇒ 00:04:15.309 Uttam Kumaran: Right, and I think you’re gonna see, like, Amber is gonna present a little bit on, like, Coco stuff this week, and she’s just a machine, so she’s pushing a lot of that out. She’s gonna push a playbook on… for everybody on, like, how to create and publish semantic views.

38 00:04:15.380 ⇒ 00:04:20.640 Uttam Kumaran: And then I think her and Kai are gonna be able to just attack the Power BI

39 00:04:20.820 ⇒ 00:04:25.720 Uttam Kumaran: problem, and that’s the directive I kind of gave her, is like, There’s a clear directive.

40 00:04:26.100 ⇒ 00:04:30.400 Uttam Kumaran: To… for everybody to be able to at least do everything that they were doing there.

41 00:04:31.060 ⇒ 00:04:38.620 Uttam Kumaran: And, of course, plus, plus, plus all the features, you know? So, I think she has, like, a pretty clear, clear mission.

42 00:04:38.930 ⇒ 00:04:49.450 Katherine Bayless: Nice, yeah. Like, I think, yeah, I think you’re right. Like, I think I will be kind of done needing to do the convincing politicking, and now I can switch to the prioritization politicking.

43 00:04:49.450 ⇒ 00:04:51.569 Uttam Kumaran: Yeah. That’s good.

44 00:04:51.570 ⇒ 00:05:05.249 Katherine Bayless: But actually, similar to the, you know, Kai and Amber can pair on the semantic views, the new data engineer is starting Monday, so Kyle’s going out for his initial two weeks parental leave, probably by Wednesday.

45 00:05:05.250 ⇒ 00:05:16.670 Katherine Bayless: So my thought is Chris, who’s the new guy, he will need to learn all the things, and also we will need somebody to keep building all the things, so I’m probably just gonna throw him at you guys.

46 00:05:16.980 ⇒ 00:05:17.570 Uttam Kumaran: Perfect.

47 00:05:17.570 ⇒ 00:05:28.450 Katherine Bayless: Yeah, and then when Kyle comes back, I want to put him in the seat that I’ve been in, which is, like, okay, like, I’m gonna hand over Fivetran to him, and, like, I want him to do the landing, and then the, like.

48 00:05:28.490 ⇒ 00:05:42.910 Katherine Bayless: QAing from the, like, you know, data engineer perspective. We’ll still have the teams QA, the, like, the sense and meaning of the data, right? You know, we still want membership to tell us that that’s the right count, but he can opine on, like, this is the right direction that we need to be going in, that kind of thing.

49 00:05:43.390 ⇒ 00:05:59.980 Katherine Bayless: And so, yeah, I think that’ll take a lot off of my plate. I think it’ll also, like, kind of taking him out of some of the engineering weeds will free him up to then… because, like, he started building, like, Claude Co- or, excuse me, Claude co-work, like, onboarding skills and stuff like that. So, like, as we start rolling that out… Right.

50 00:05:59.980 ⇒ 00:06:07.910 Katherine Bayless: He’s no longer the guy who will, like, run analysis for you, he’s the guy who’ll help you get set up to do it yourself in a very supported kind of way, so…

51 00:06:07.910 ⇒ 00:06:15.999 Uttam Kumaran: And I think this quarter has been him also learning, like, how to do it himself. That makes a lot of sense. Yeah, I mean, I also think, like.

52 00:06:16.260 ⇒ 00:06:20.219 Uttam Kumaran: Our ability to, like, get the foundation is gonna allow you to, like.

53 00:06:20.390 ⇒ 00:06:27.169 Uttam Kumaran: make a strategic political decision on, like, something new came up, like, should we take it?

54 00:06:27.170 ⇒ 00:06:28.410 Katherine Bayless: Yeah, exactly.

55 00:06:28.410 ⇒ 00:06:33.060 Uttam Kumaran: Versus, like, I don’t even have time to talk them through this idea.

56 00:06:33.170 ⇒ 00:06:33.829 Katherine Bayless: And so that.

57 00:06:33.830 ⇒ 00:06:38.119 Uttam Kumaran: That’s why even on our team, I think matching Amber and Kai together, I think Awash.

58 00:06:38.560 ⇒ 00:06:49.239 Uttam Kumaran: matching, and then I think today we’ll kind of talk… I mean, I think one is, like, today I want to talk about, like, how I can get leveraged better, and then also I want to talk about, like.

59 00:06:49.650 ⇒ 00:06:59.170 Uttam Kumaran: as this grows, and as the relationship grows, one thing I have in the kind of proposal is, like, maybe we need to bring on someone from our side that’s more of, like.

60 00:06:59.280 ⇒ 00:07:12.910 Uttam Kumaran: kind of, like, more program management, or, like, the sprint-by-sprint, like, organization. Like, we… on our side, like, everybody on our side is, like, technical, but we have some people that skew more, like, they’re, like, organization, like.

61 00:07:13.120 ⇒ 00:07:18.120 Uttam Kumaran: psychos, right? And so they’re, like, really good at, like, doing that.

62 00:07:18.120 ⇒ 00:07:27.820 Uttam Kumaran: And they’re, like, very opinionated about, like, the Gantt and the view and, like, how to manage up, so it’s, like, at any moment, if Catherine needs

63 00:07:27.820 ⇒ 00:07:39.020 Uttam Kumaran: like, the Gantt chart, or if you need, like, what the roadmap is, or if there’s, like, another executive that, like, needs to get an email formatted in a certain way, you can just be like, hey guys, I’m talking to this person.

64 00:07:39.680 ⇒ 00:07:45.889 Uttam Kumaran: I’m like, I can’t… I need just, like… can we just get me an email blurb, like, about our project? That’s…

65 00:07:46.020 ⇒ 00:07:58.879 Uttam Kumaran: But we have some other clients where we’re supporting in that way, right? So that way, you can go meeting and be like, hey guys, I just… I had this great in on this project, I just need a couple of materials to support, and like…

66 00:07:59.230 ⇒ 00:08:05.179 Uttam Kumaran: And then as well as, like, okay, now we’re gonna have, like, 5, 6, 7 work streams.

67 00:08:05.600 ⇒ 00:08:06.560 Uttam Kumaran: Like…

68 00:08:06.560 ⇒ 00:08:07.580 Katherine Bayless: So, those…

69 00:08:07.580 ⇒ 00:08:16.990 Uttam Kumaran: Those are the two things, which is, like, managing up on, like, how does our team come across to the org really organized? And then the second is, like, literally the week-to-week,

70 00:08:17.610 ⇒ 00:08:22.670 Uttam Kumaran: Like, the stand-ups are running, each team, like, has clarity on, like, what the objectives are, so…

71 00:08:22.910 ⇒ 00:08:27.060 Katherine Bayless: Yeah, and then I think the third value that’ll come out of that is

72 00:08:27.470 ⇒ 00:08:40.079 Katherine Bayless: If we have this, and everybody can see it, as we start to grow, you know, sort of, like, technical skills, like, on the local teams, like, if they see a story on the board, like, they can grab it, right?

73 00:08:40.080 ⇒ 00:08:40.720 Uttam Kumaran: Yes.

74 00:08:40.720 ⇒ 00:08:47.800 Katherine Bayless: somebody might be like, look, I don’t know how to, like, actually put a semantic model into GitHub, but I can go in and go ahead and do all the defining and provide all these.

75 00:08:47.800 ⇒ 00:08:49.560 Uttam Kumaran: Exactly, exactly.

76 00:08:49.560 ⇒ 00:08:50.829 Katherine Bayless: Like, I really want.

77 00:08:50.830 ⇒ 00:08:52.220 Uttam Kumaran: the context creation.

78 00:08:52.370 ⇒ 00:08:57.519 Katherine Bayless: Exactly, exactly, exactly. Yep. Yep, yep, yep. I did.

79 00:08:57.520 ⇒ 00:08:59.300 Uttam Kumaran: Pretty cool. Yeah. Oh, yeah.

80 00:08:59.300 ⇒ 00:09:07.260 Katherine Bayless: For this year, to get 5 people outside of my team into, like, a, you know, technical subject matter expert on our.

81 00:09:07.260 ⇒ 00:09:15.679 Uttam Kumaran: Oh, nice! Yeah, I mean, if they can be the bridge between their business unit and us, then we’ll start to just pull them into our…

82 00:09:16.570 ⇒ 00:09:26.059 Uttam Kumaran: So I’m doing that with our analyst team. Like, Amber, for example, I meet with our… the people on, sort of, our analyst service, and people are like, well.

83 00:09:26.090 ⇒ 00:09:46.089 Uttam Kumaran: because I was talking about, like, we’re coming up with these skills to interact with DPT, and they’re like, yeah, it’s difficult, because sometimes I’m, like, it’s hard for me, like, one of the things that’s been challenging is, like, I have to go into Cursor or something and find out, like, the right model to use, or find out the column that I need to change, but then I throw it to the AE team, and, like, they need to run, I’m like.

84 00:09:46.120 ⇒ 00:09:59.370 Uttam Kumaran: well, actually, like, you guys can go make that change, and I think I’ll show you how to write skills to make those, because then, ultimately, you can make the change, test that it runs, and then push the PR and be like, hey guys, I needed this, I made this change.

85 00:09:59.580 ⇒ 00:10:08.659 Uttam Kumaran: And then the AE who’s typically there, is able to go work on, like, the really intense, like, cohorting model or some, like, runtime issue, like…

86 00:10:08.730 ⇒ 00:10:22.590 Uttam Kumaran: Versus, like, changing a column name. But it’s funny, because they’re… they’re like, I don’t want to break anything. I’m like, you won’t break anything. It’s… so, like, that’s… that’s, like, on the platform, or whatever platform, to prevent you from breaking anything.

87 00:10:22.590 ⇒ 00:10:30.410 Uttam Kumaran: I’m, like, giving you the dev schema and the instructions so that your agent knows, if I’m making a change I want to test, do it in dev locally, right?

88 00:10:30.470 ⇒ 00:10:40.260 Uttam Kumaran: And that way, I told them, you’re not gonna break anything. In fact, I want to give you more power so that you can work with a client, come back, and in the same day, flip

89 00:10:40.290 ⇒ 00:10:57.179 Uttam Kumaran: around a column name change, a slight case when change, versus, like, submitting a ticket, having it go there, that person doesn’t do it until next week, they’re out of office, right? Like, so we’re… so I’m pulling them even further, which has been fun. Which has been fun.

90 00:10:57.180 ⇒ 00:10:59.610 Katherine Bayless: Like, it’s exactly where my head goes with all the time.

91 00:10:59.610 ⇒ 00:11:10.079 Uttam Kumaran: I did it without anybody pulling me, because I was like, I just… I’m like, I’m gonna fish, this is taking way too long, I’m just gonna learn this technology. Exactly! I mean, right? I know skills work.

92 00:11:10.080 ⇒ 00:11:12.920 Katherine Bayless: No, I haven’t degree, yeah, exactly.

93 00:11:14.240 ⇒ 00:11:17.149 Katherine Bayless: I just solved problems really well.

94 00:11:17.150 ⇒ 00:11:18.480 Uttam Kumaran: Yeah, yeah.

95 00:11:18.800 ⇒ 00:11:22.830 Katherine Bayless: But yeah, I think that’s exactly where we’re gonna be able to get to, which is cool.

96 00:11:23.370 ⇒ 00:11:23.930 Uttam Kumaran: Cool.

97 00:11:24.540 ⇒ 00:11:25.119 Katherine Bayless: But yeah.

98 00:11:25.120 ⇒ 00:11:26.939 Uttam Kumaran: I actually… yeah, yeah, go ahead.

99 00:11:26.940 ⇒ 00:11:28.009 Katherine Bayless: No, no, go ahead.

100 00:11:28.370 ⇒ 00:11:37.770 Uttam Kumaran: Yeah, so I put together… I sort of took your feedback on the doc, and I just put together, another version, and I think it… this is sort of aligns with.

101 00:11:37.770 ⇒ 00:11:38.100 Katherine Bayless: Okay.

102 00:11:38.100 ⇒ 00:11:42.779 Uttam Kumaran: What, what, we talked about, and it’s funny because…

103 00:11:43.810 ⇒ 00:11:56.589 Uttam Kumaran: my business partner, Robert, he’s a lot more on the commercial term side of the business, and I was like, hey, we have, you know, Catherine at CTA is interested in us moving away from sort of outcomes-focused, but sort of, like.

104 00:11:56.720 ⇒ 00:12:08.760 Uttam Kumaran: fractional team as a service that we can kind of scale up and down. And we actually have another client that’s doing exactly this, where we basically have… we’re on retainer with them, and we handle, like.

105 00:12:08.950 ⇒ 00:12:17.390 Uttam Kumaran: tons of new requests that come in, and then in certain events where the scope really changes, we sort of have, like, breakers, where we’re like, hey, this is, like.

106 00:12:17.540 ⇒ 00:12:19.520 Uttam Kumaran: Pretty much a net new program.

107 00:12:19.640 ⇒ 00:12:20.079 Katherine Bayless: That we would.

108 00:12:20.080 ⇒ 00:12:30.910 Uttam Kumaran: need some external… extended budget for. And so, really, like, what this scope… what this document is… is really just, like, highlighting, one is, like, it’s a single agreement.

109 00:12:30.910 ⇒ 00:12:41.650 Uttam Kumaran: like, we wanted to structure it as a 12-month agreement, but have quarterly sort of alignments for us to say, hey, like, what is the priorities for the quarter? Because I still am, like.

110 00:12:42.100 ⇒ 00:12:51.870 Uttam Kumaran: I try to align as much towards, like, the outcomes that we’re driving towards, but I’ve totally heard with… this is the second quarter we kind of tried, and still things floated.

111 00:12:51.970 ⇒ 00:12:59.059 Uttam Kumaran: quite a bit, you know? So that’s… that’s fine, and I think what we tried to do here is weave in, like, your…

112 00:12:59.160 ⇒ 00:13:07.690 Uttam Kumaran: percent of effort sort of situation, and give, like, sample items on, like, what could be some deliverables.

113 00:13:07.690 ⇒ 00:13:08.230 Katherine Bayless: Yeah.

114 00:13:08.230 ⇒ 00:13:12.970 Uttam Kumaran: So it’s clear, like, What that work stream could include.

115 00:13:13.200 ⇒ 00:13:13.690 Katherine Bayless: Yeah.

116 00:13:13.690 ⇒ 00:13:27.399 Uttam Kumaran: And so, like, that’s sort of, like, the copy at the top here, and then we sort of break into, like, okay, there’s, like, data ingestion, there’s, like, semantic layer, there’s infrastructure, there’s support and adoption, and then I sort of mentioned some of this, like.

117 00:13:28.930 ⇒ 00:13:34.960 Uttam Kumaran: there may be in… there may be some of these short AI engineering activities, and then maybe that

118 00:13:35.120 ⇒ 00:13:41.529 Uttam Kumaran: we, like, basically I said, hey, if this ends up being something larger, then we scope it out and it goes larger.

119 00:13:41.530 ⇒ 00:13:41.920 Katherine Bayless: Yeah.

120 00:13:41.920 ⇒ 00:13:53.319 Uttam Kumaran: Which is, like, why I wanna… I wanna get this through the line and then call Jay. And this is where, like, I want to also have the freedom to be like, call Jay. Here, if it’s, like, small things, like, we’ll just… we’ll just assist and do that.

121 00:13:53.590 ⇒ 00:13:53.930 Katherine Bayless: Yeah.

122 00:13:53.930 ⇒ 00:14:00.630 Uttam Kumaran: If it ends up being, like, okay, this is a larger thing, then that’s how we’ll do it. So that’s sort of how we architected this.

123 00:14:01.240 ⇒ 00:14:16.510 Katherine Bayless: I think absolutely perfect. Totally read my mind. Two things. One is, and it’s gonna be… it’s gonna sound contradictory, but, I think this is the right direction to go in. I did suggest it to Christina, and she did not like it.

124 00:14:16.510 ⇒ 00:14:16.850 Uttam Kumaran: Okay.

125 00:14:16.850 ⇒ 00:14:35.409 Katherine Bayless: Oh, a retainer? No, that makes me very nervous. Okay. Mmm, sounds real convenient. But I think that there’s a way for us to… I think if we put in the Q2 scope with this structure, and we say, like, it’s similar to a retainer, but to your point, there’s, like, a kind of…

126 00:14:35.410 ⇒ 00:14:37.239 Uttam Kumaran: That’s exactly what we tried to do here.

127 00:14:37.240 ⇒ 00:14:37.680 Katherine Bayless: Exactly.

128 00:14:37.680 ⇒ 00:14:41.860 Uttam Kumaran: is I put, like, here’s the fir- here are the objectives for the first quarter window.

129 00:14:41.860 ⇒ 00:14:42.230 Katherine Bayless: Yep.

130 00:14:42.230 ⇒ 00:14:58.059 Uttam Kumaran: And moving into every quarter, what you can expect from us is an alignment on, like, what is our objective for the next quarter. Certainly, it’s fair expectation, and this is where I also agree, I don’t like retainers, even for our business, because I think it gets… it makes…

131 00:14:58.250 ⇒ 00:15:08.050 Uttam Kumaran: And look, I think we’re all, like, I can vouch we’re good stewards of the budget, but I can see that from her perspective, like, I think that’s totally fair. And so one is, like, I think…

132 00:15:08.310 ⇒ 00:15:21.769 Uttam Kumaran: at minimum, these are the objectives that I listed, which I think we should just align on. Are these the things that we feel like we want to take on this quarter? And I… I see to this a little bit, but that’s this… that’s…

133 00:15:21.770 ⇒ 00:15:28.410 Uttam Kumaran: that’s my objective today, is to make sure of this. And then, a lot of the language in the agreement is about, like.

134 00:15:28.450 ⇒ 00:15:31.110 Uttam Kumaran: This is structured as, like, this retainer.

135 00:15:31.430 ⇒ 00:15:46.060 Uttam Kumaran: it’s structured kind of as a fractional team, but ultimately, every quarter, we are driving towards objectives, you know? And we’re using the budget. All of these things take percentages of those pieces, and

136 00:15:46.580 ⇒ 00:15:52.680 Uttam Kumaran: the alternative is to do, basically, scopes for every single thing, and I think

137 00:15:52.920 ⇒ 00:15:57.420 Uttam Kumaran: I’m sure she’s, like, that’s… things are changing pretty fast, you know?

138 00:15:57.420 ⇒ 00:16:17.329 Katherine Bayless: Exactly, yeah, totally. And so I’m like, I think what we can do is basically, like, you know, demonstrate that this will work with this quarter, and then the next one we put in goes for… actually, I do think we will have to still do 6 months, and then break. Okay. We could do a year, because I don’t think finance allows multi-year contracts.

139 00:16:18.750 ⇒ 00:16:20.309 Uttam Kumaran: What’s a year in terms of calendar year, or in terms.

140 00:16:20.310 ⇒ 00:16:38.570 Katherine Bayless: Yeah, like, they don’t like it to cross years. So we might get stuck with two more, but I do want to move this direction, so I think it’s, like, we use this as our, like, example of, like, this is what we mean when we say, like, we want to kind of make it a little bit more flexible and a little bit more, like, you know, expandable, because I…

141 00:16:38.570 ⇒ 00:16:38.970 Uttam Kumaran: Yeah.

142 00:16:39.110 ⇒ 00:16:50.000 Katherine Bayless: to your point, we want to have a ceiling so that we don’t go crazy, but also, like, if every new idea has to come through and, like, you know, go through a scoping thing, then it’s like, okay, well, now we’re not really getting the…

143 00:16:50.000 ⇒ 00:16:53.560 Uttam Kumaran: Yeah. Exactly. And again, like, this is where, like.

144 00:16:53.660 ⇒ 00:16:58.170 Uttam Kumaran: to tell you, as a vendor, like, I have historically… we will just…

145 00:16:58.310 ⇒ 00:17:04.949 Uttam Kumaran: we will always be like, let’s just do the work. Like, I’ve… it’s so funny, because I’ve… I’ve… we’ve always been like that, just because…

146 00:17:04.990 ⇒ 00:17:19.280 Uttam Kumaran: I’m not used to op… I’m not used to operating as, like, a, like, stern consultant, where we’re like, oh, no, no, we have to, like, sign. I’m actually much more in the other camp, but it’s not good for business to do that. So, like, I think that’s…

147 00:17:19.280 ⇒ 00:17:28.749 Uttam Kumaran: this is… I think this is basically the second and third section. Maybe you can let me know how… or it’s basically, like, section 1, 2, and 3.

148 00:17:28.960 ⇒ 00:17:29.350 Katherine Bayless: It’s like.

149 00:17:29.350 ⇒ 00:17:32.860 Uttam Kumaran: the package for Christina that I just want to nail.

150 00:17:33.100 ⇒ 00:17:34.060 Uttam Kumaran: Yeah.

151 00:17:34.060 ⇒ 00:17:34.600 Katherine Bayless: Yeah.

152 00:17:34.600 ⇒ 00:17:40.980 Uttam Kumaran: You know, so maybe we can work through that today, but maybe if I… I’ll just… let me just walk through the rest of the docs that you have.

153 00:17:40.980 ⇒ 00:17:41.469 Katherine Bayless: I’m delayed.

154 00:17:41.470 ⇒ 00:17:42.770 Uttam Kumaran: Play the land, and then…

155 00:17:42.950 ⇒ 00:17:48.499 Uttam Kumaran: So really, this is just, like, what are some potential deliverables on each of the work streams?

156 00:17:50.320 ⇒ 00:17:58.989 Uttam Kumaran: So this is… these are the work streams that I sort of came up with, which is, like, there’s data ingestion, there’s dbt, there’s all the… anything around semantic layer AI context.

157 00:17:59.160 ⇒ 00:18:09.899 Uttam Kumaran: like, we put Cortex, you know, you can assume it could be any other surface, really, but, some of these… that whole area is here. Infrastructure.

158 00:18:10.400 ⇒ 00:18:24.530 Uttam Kumaran: Which is just, like, anything around cleanup or, like, broader protection beyond just, like, data. We have, like, user support and adoption, which I wanted to carve out in particular for, like, training, support, and then this is just where it’s, like.

159 00:18:24.810 ⇒ 00:18:26.519 Uttam Kumaran: Okay, if some of these are just, like.

160 00:18:26.680 ⇒ 00:18:37.189 Uttam Kumaran: meetings and consulting, or then it’s like… or some of these become, like, their own workstreams entirely. I’ve sort of talked about, which is just, like.

161 00:18:37.620 ⇒ 00:18:40.699 Uttam Kumaran: This bucket is not part of the split, unless it’s, like.

162 00:18:41.200 ⇒ 00:18:44.630 Uttam Kumaran: We… we find that it’s… it’s actually worth

163 00:18:44.790 ⇒ 00:18:47.009 Uttam Kumaran: Pulling out into its own item.

164 00:18:47.230 ⇒ 00:18:52.690 Uttam Kumaran: And then I… I kind of talk about some phasing, but I think this is where it’s… this is just, like.

165 00:18:53.320 ⇒ 00:19:01.610 Uttam Kumaran: as we’re going out through the weeks, so this is more illustrative of, like, what it could look like. I can even remove this if it’s, like, not worth putting this in here.

166 00:19:02.410 ⇒ 00:19:04.619 Katherine Bayless: I mean, it’s funny, like, to your point, like.

167 00:19:05.090 ⇒ 00:19:14.899 Katherine Bayless: in a, like, aspirational sense, right? Like, I love this. I mean, I wish every company wanted to contract like this, but then it’s like, I feel like the reality we’re living is.

168 00:19:14.900 ⇒ 00:19:15.660 Uttam Kumaran: Yeah.

169 00:19:15.660 ⇒ 00:19:20.269 Katherine Bayless: It was like, we’re coupling a planning process with a contracting process, which is kind of clunky.

170 00:19:20.270 ⇒ 00:19:20.740 Uttam Kumaran: Correct.

171 00:19:20.740 ⇒ 00:19:23.050 Katherine Bayless: It’s just changing too fast, right?

172 00:19:23.050 ⇒ 00:19:26.510 Uttam Kumaran: Yeah, so that’s why it’s like, I think for some of our clients.

173 00:19:26.760 ⇒ 00:19:35.980 Uttam Kumaran: I hesitate to put this in, because some of them will come back and, like, at the end, someone, like, a Christina will be like, well, you didn’t deliver on this, and then I’m like.

174 00:19:36.220 ⇒ 00:19:46.980 Uttam Kumaran: But we… we did… we did so much other stuff, they’re like… and then their whole context, so I would prefer… if this isn’t helping anyone, I would prefer to leave it out, because I think…

175 00:19:47.310 ⇒ 00:19:51.489 Uttam Kumaran: the rest of this sort of hedges, so if you’re okay with that, I also.

176 00:19:51.490 ⇒ 00:19:51.810 Katherine Bayless: agree.

177 00:19:51.810 ⇒ 00:20:05.990 Uttam Kumaran: in that, like, I really want to be able to look at this scope at the end of the quarter and been like, we were spot on. Yeah. Right? And so that’s what I felt in the last scope. For some of those, we hit, some of those that got pushed or deprioed, and, like.

178 00:20:06.200 ⇒ 00:20:12.680 Uttam Kumaran: from a Christina lens, or the business lens, that may be the only artifact Apart from, like.

179 00:20:12.950 ⇒ 00:20:19.850 Uttam Kumaran: the team that they have, and like, I just want to make sure that anyone can look at that and be like, yeah, check, check, check, like, we’re…

180 00:20:20.130 ⇒ 00:20:21.680 Uttam Kumaran: We feel good, so…

181 00:20:21.680 ⇒ 00:20:41.039 Katherine Bayless: completely agree. I mean, I think also, like, some of this stuff winds up, like, getting, like, made moot. Like, it’s funny, in between Kai drafting her goals and me finally, like, packaging them up to go to the next level for review, like, one of her goals was, like, migrate at least 10 Power BI reports, and I was like, oh, that’s actually just… that’s just all done now.

182 00:20:41.040 ⇒ 00:20:41.640 Uttam Kumaran: Yeah.

183 00:20:41.640 ⇒ 00:20:44.680 Katherine Bayless: Like, you’ve got till December to do something we’ve already finished, basically.

184 00:20:44.680 ⇒ 00:20:45.700 Uttam Kumaran: Yeah, yeah.

185 00:20:45.700 ⇒ 00:20:49.329 Katherine Bayless: You know? And so it’s like, the work kind of disappears out from under us in some ways.

186 00:20:49.330 ⇒ 00:20:56.640 Uttam Kumaran: Yeah, that’s what’s really hard. And again, I think you’re hitting it on the head, which is, like, typical businesses…

187 00:20:56.970 ⇒ 00:21:07.419 Uttam Kumaran: like, typical enterprises, they don’t couple this together. It’s sort of like, we put these scopes, and then there’s the planning, and then they, like, wrap it up, but I don’t know, how…

188 00:21:07.480 ⇒ 00:21:25.470 Uttam Kumaran: at our pace, like, there’s some other clients of ours where I’m like, okay, they wanna go, they wanna do the normal thing, and so we’re gonna be really specific on what we’re not doing, and then anything has to go through change order, because I know that they’re gonna look back at this SOW. In this case, I’m like, hey, if… if, like.

189 00:21:25.570 ⇒ 00:21:34.669 Uttam Kumaran: snow work comes out in a big way, and that cuts this, we better… we jump on that, like, and that’s in the best interest of CTA, you know?

190 00:21:34.670 ⇒ 00:21:36.739 Katherine Bayless: Exactly. Yeah.

191 00:21:36.740 ⇒ 00:21:45.949 Uttam Kumaran: And that’s also taking advantage of us, because that’s… that’s, I feel like, why we’re different, in that, like, we’re not like most vendors where we’re like, oh, we don’t have anyone trained on that. I’m like.

192 00:21:46.660 ⇒ 00:22:01.619 Uttam Kumaran: I’ll call Amber, like, let’s get… like, let’s fix… like, can we just do, like, a two-hour working session, see if we can get this set up? Okay, we generally have the understanding, let’s run. So, I also think, like, our clients should take advantage of us in that way, that, like, that’s how we…

193 00:22:01.880 ⇒ 00:22:03.370 Uttam Kumaran: We do things, you know.

194 00:22:03.590 ⇒ 00:22:06.210 Katherine Bayless: Yeah, no, completely agree. Completely agree.

195 00:22:06.210 ⇒ 00:22:13.840 Uttam Kumaran: But it’s tough, because, like, it’s hard… like, when you’re just looking at this piece of paper, I could see how, like, yeah, it’s a little bit different.

196 00:22:13.980 ⇒ 00:22:19.950 Uttam Kumaran: So this is just, like, risks. Again, I think this is just, like, writing some of them down on, like.

197 00:22:21.400 ⇒ 00:22:37.720 Uttam Kumaran: some of the risks that we’ve outlined before. We talked about, like, acceptance criteria. I basically said it tries to focus on these outcomes by the end of the quarterly window. Again, this is a kind of a corollary to three, so if you prefer, like, this

198 00:22:37.830 ⇒ 00:22:43.669 Uttam Kumaran: I was like, I didn’t… I didn’t… I think it’s a little bit redundant. I could just consolidate it all into one.

199 00:22:44.140 ⇒ 00:22:44.690 Katherine Bayless: So…

200 00:22:44.950 ⇒ 00:22:50.270 Uttam Kumaran: These are the… maybe we can even just talk about it. These are some of the ones that we try to…

201 00:22:50.520 ⇒ 00:22:54.429 Uttam Kumaran: Be like, okay, in this quarter, we… we want to… we want to do these.

202 00:22:55.740 ⇒ 00:23:05.170 Katherine Bayless: Yeah, I mean, I think, yeah, like, however we consolidate the language, these are definitely, like, the things that we want, so… yeah.

203 00:23:05.170 ⇒ 00:23:09.919 Uttam Kumaran: Okay. Yeah, and again, I think it’s like, if we look between now and the end of the quarter, these seem like

204 00:23:10.310 ⇒ 00:23:15.210 Uttam Kumaran: Unless, literally, like, something dramatic changes, these seem pretty, like, fair.

205 00:23:15.420 ⇒ 00:23:30.999 Katherine Bayless: Yeah, I mean, I think really the… I mean, to your point, it could be functionality or political stuff on this side. Like, functionality-wise, yes, things could change and accelerate. Politically, I mean, the change in the acceleration is, like, everybody wants their data in Snowflake.

206 00:23:31.000 ⇒ 00:23:31.440 Uttam Kumaran: Okay.

207 00:23:31.440 ⇒ 00:23:40.800 Katherine Bayless: do think we are gonna start landing a lot more data. I also think a lot of the stuff that people are interested in pulling in next is, like.

208 00:23:41.010 ⇒ 00:23:46.789 Katherine Bayless: the type of things that are sort of more, like, garden variety, where, like, you guys won’t necessarily need to understand, like, the CTA context.

209 00:23:46.790 ⇒ 00:23:47.519 Uttam Kumaran: Yeah, yeah, yeah, yeah.

210 00:23:47.520 ⇒ 00:23:52.910 Katherine Bayless: We want to bring in all of our paid advertising data, all of our social media stuff, right? And that’s, you know, data engineering…

211 00:23:52.910 ⇒ 00:24:00.479 Uttam Kumaran: fairly standard stuff on our side, yeah, so we’ve done anything that’s classic, like paid marketing, organic marketing.

212 00:24:00.870 ⇒ 00:24:04.930 Uttam Kumaran: Like, traditional ClickFunnel analytics, traditional…

213 00:24:05.190 ⇒ 00:24:10.339 Uttam Kumaran: sales, refunds, discounts, stuff like that. It’s, it’s, like, that’s all we do, you know?

214 00:24:10.340 ⇒ 00:24:18.220 Katherine Bayless: Yeah. And so, like, if anything, maybe that’s really the one that you could change would be for CES analytics, like, maybe that’s just more, like.

215 00:24:19.880 ⇒ 00:24:32.360 Uttam Kumaran: I wanted to say CSN, because I feel like that’s everything around, like, that excludes the other events, it kind of excludes marketing that’s more broader, so I can put something…

216 00:24:32.630 ⇒ 00:24:37.590 Uttam Kumaran: around, like, maybe scoping that, like, broader marts, which is, like.

217 00:24:37.910 ⇒ 00:24:41.979 Uttam Kumaran: CTA, paid marketing, anything like that, but I know…

218 00:24:42.100 ⇒ 00:24:46.700 Uttam Kumaran: our initial goal was to just put a pin in, like, CES 2026.

219 00:24:47.580 ⇒ 00:24:48.270 Uttam Kumaran: you know.

220 00:24:48.850 ⇒ 00:25:08.570 Katherine Bayless: Yeah, I know, it’s funny too, because, like, we’ve put out the audited report and the press release and everything, so it’s, like, now everybody’s already, like, moved on, you know? And so, like, now it’s kind of, like, all eyes on next year’s data and all the things we want to get done in the meantime, even though I, like, I do think there’s still some work for us to finish out.

221 00:25:08.950 ⇒ 00:25:15.139 Katherine Bayless: So I’m like, I wonder if that one becomes more of a generic data ingestion kind of thing, and then the.

222 00:25:15.140 ⇒ 00:25:15.790 Uttam Kumaran: Yeah.

223 00:25:15.790 ⇒ 00:25:20.320 Katherine Bayless: It’s called out as, like, A critical deliverable under that.

224 00:25:20.630 ⇒ 00:25:21.540 Katherine Bayless: broader, I don’t.

225 00:25:21.540 ⇒ 00:25:28.399 Uttam Kumaran: Maybe I could just say… I mean, I would just think about, like, the outcome is, like, CTA data marts, right? And so these could include, like.

226 00:25:28.650 ⇒ 00:25:31.510 Uttam Kumaran: marketing,

227 00:25:31.940 ⇒ 00:25:39.070 Uttam Kumaran: like, broader sales, like, the site analytics, right? I don’t know, is there anything else that’s, like.

228 00:25:40.450 ⇒ 00:25:41.080 Katherine Bayless: to march.

229 00:25:41.080 ⇒ 00:25:42.040 Uttam Kumaran: burger. Yeah.

230 00:25:42.040 ⇒ 00:25:56.370 Katherine Bayless: We’re gonna… I think we’re gonna get our financials, contracts, we… the enrichment vendors, or the enrichment stuff’s gonna happen, we’ve got memberships to give us the API key to sales intel, I think there was one other…

231 00:25:57.440 ⇒ 00:25:59.820 Katherine Bayless: Big picture category.

232 00:26:02.810 ⇒ 00:26:06.919 Katherine Bayless: No. Maybe that covers it.

233 00:26:07.290 ⇒ 00:26:07.870 Katherine Bayless: Well, okay.

234 00:26:07.870 ⇒ 00:26:09.000 Uttam Kumaran: Yeah, Marcus?

235 00:26:09.000 ⇒ 00:26:18.520 Katherine Bayless: So under the marketing one, it’s the, like, your paid ads and stuff like that, but then they also were going to bring in the data from TalkWalker, which is, like, just kind of general social monitoring type stuff.

236 00:26:18.520 ⇒ 00:26:18.880 Uttam Kumaran: Yeah.

237 00:26:18.880 ⇒ 00:26:19.779 Katherine Bayless: So, yeah.

238 00:26:20.650 ⇒ 00:26:25.879 Uttam Kumaran: Okay, so I can put something that’s more generic on this. I mean, I think what I’ll put here is that, like.

239 00:26:26.730 ⇒ 00:26:29.580 Uttam Kumaran: Yeah, we’re sort of… it’s the ingestion around…

240 00:26:29.740 ⇒ 00:26:32.330 Uttam Kumaran: this, and then I think probably next quarter…

241 00:26:32.590 ⇒ 00:26:39.729 Uttam Kumaran: Or maybe around June is when we start, like, modeling, you know, a lot of this, because then that’ll feed into, like, all of these.

242 00:26:40.170 ⇒ 00:26:41.590 Uttam Kumaran: surfaces, you know?

243 00:26:41.590 ⇒ 00:26:50.519 Katherine Bayless: Right, yeah, because, I mean, honestly, like, what people are, like, you know, if we think about, like, what are the behaviors people are engaging in, like, right now, it’s a lot of, you know.

244 00:26:50.520 ⇒ 00:27:04.830 Katherine Bayless: program analysis and planning, like, you know, what do we want to do or change for next year kind of conversations. But I think, I mean, yeah, definitely, as soon as we get to July and the Innovation Awards start running, like, that’s kind of, like, the first, like.

245 00:27:04.920 ⇒ 00:27:06.790 Katherine Bayless: You know, sort of, like, opener…

246 00:27:06.790 ⇒ 00:27:07.210 Uttam Kumaran: season.

247 00:27:07.210 ⇒ 00:27:22.169 Katherine Bayless: ES stuff, and it’s gonna immediately search to, like, a switch to that telemetry, right? Like, how are we doing? Who’s talking about what? How are those ads performing? What are the exhibits? Like, it’s… yeah, it’s gonna go so fast from, like, analytics to telemetry, it’ll snap your neck.

248 00:27:22.170 ⇒ 00:27:28.020 Uttam Kumaran: Yeah, so then this, I think, is everything around ingestion this quarter for these, as much as possible.

249 00:27:28.020 ⇒ 00:27:28.870 Katherine Bayless: Exactly.

250 00:27:28.870 ⇒ 00:27:33.680 Uttam Kumaran: I mean, as soon as it’s landed, we will move to model, you know, as fast as we can.

251 00:27:33.900 ⇒ 00:27:35.400 Katherine Bayless: Yeah, yeah.

252 00:27:35.690 ⇒ 00:27:36.630 Uttam Kumaran: Yeah. Okay.

253 00:27:37.130 ⇒ 00:27:38.940 Katherine Bayless: Yeah, I like it, though.

254 00:27:39.750 ⇒ 00:27:49.659 Uttam Kumaran: And then on our team, I basically had, I just… the only changes really here is, like, I listed Robert, who leads strategy and analytics, like, on our side.

255 00:27:49.790 ⇒ 00:27:54.509 Uttam Kumaran: I think, like, Amber is sort of under that service. I think.

256 00:27:54.620 ⇒ 00:28:06.389 Uttam Kumaran: I’m just gonna leave him here for now, because I… I think as… in case we have specific, like, questions about certain strategy, like, I think he’s… he’s gonna be the one to sort of fill in, but his team…

257 00:28:06.490 ⇒ 00:28:20.319 Uttam Kumaran: is pretty good. And then I just had one listing here for, like, someone around delivery, so I think maybe we can answer this question, like, do you think this is needed sooner than later? Like, there… I… we have a great person who just joined our team recently who’s, like.

258 00:28:20.600 ⇒ 00:28:30.729 Uttam Kumaran: this is, like, that he’s that guy. So I wonder if you think this is, like, more of a this-quarter thing, or if we just should just continue to focus on the way we’re going, like.

259 00:28:31.050 ⇒ 00:28:32.220 Uttam Kumaran: What do you think?

260 00:28:32.550 ⇒ 00:28:36.289 Katherine Bayless: I know they’re, like, Christina’s very eager to…

261 00:28:36.290 ⇒ 00:28:54.110 Katherine Bayless: test the internal ability to, like, really get that engine going, because we haven’t really tried, organizationally, to, like, have a, you know, cohesive thought around project management, so I and her work, and then Jay’s got somebody coming in on his team, I think, Monday as well, who’s going to be an analyst for IT, and so…

262 00:28:54.110 ⇒ 00:28:54.670 Uttam Kumaran: Okay.

263 00:28:54.670 ⇒ 00:29:01.000 Katherine Bayless: I think we want to kind of give them a chance to get things moving, That being said.

264 00:29:01.000 ⇒ 00:29:02.630 Uttam Kumaran: But that’s, like, company-wide?

265 00:29:02.630 ⇒ 00:29:05.930 Katherine Bayless: Yeah, exactly. So it’d be, like, trying to see if we can push this out.

266 00:29:05.930 ⇒ 00:29:06.450 Uttam Kumaran: Okay.

267 00:29:06.450 ⇒ 00:29:10.219 Katherine Bayless: If this person could be available, maybe as, like, you know, a light.

268 00:29:10.220 ⇒ 00:29:10.660 Uttam Kumaran: Again, yeah.

269 00:29:10.660 ⇒ 00:29:21.610 Katherine Bayless: light touch where Kai can reach out to them for, like, mentorship and thought leadership type stuff? Not a bad thing. And then that way, too, if we start to realize, like, yeah, no, we’re gonna need some horsepower to keep up with this.

270 00:29:21.610 ⇒ 00:29:30.960 Katherine Bayless: then it’s like, I happen to know somebody we could bring in, and they’ve already got a little bit of context, you know, that kind of thing. But I do think Christina is looking to see if we can

271 00:29:31.600 ⇒ 00:29:35.519 Katherine Bayless: You know, handle our own on the project management front to a certain extent.

272 00:29:35.810 ⇒ 00:29:52.200 Uttam Kumaran: Okay, cool, then maybe, like, and then kind of, like, what are your expectations for Kai on, like, the sort of coordination project management side? Because that way I can… Garrett is on our team, I can share that with him, and then make the connection, and just make sure Kai feels supported.

273 00:29:53.020 ⇒ 00:30:09.469 Katherine Bayless: Yeah, I mean, I think there’s still a lot of TBDs, right? I mean, everything you were talking about earlier in terms of, like, first thing is we need to get our stuff in order so that, you know, everybody can kind of see where everything is, then we want to graduate to, like, okay, and if you want to contribute, grab a story. And so, like, Kai’s in that headspace right now.

274 00:30:09.470 ⇒ 00:30:20.239 Katherine Bayless: Trying to figure out, like, what does this Asana board need to look like? But I think her having somebody she could bounce ideas off of would be really beneficial.

275 00:30:20.360 ⇒ 00:30:21.040 Katherine Bayless: But yeah.

276 00:30:21.040 ⇒ 00:30:21.730 Uttam Kumaran: Okay.

277 00:30:21.730 ⇒ 00:30:22.320 Katherine Bayless: Yeah.

278 00:30:22.320 ⇒ 00:30:23.130 Uttam Kumaran: Okay, cool.

279 00:30:23.490 ⇒ 00:30:26.480 Katherine Bayless: And then I think also, probably, like, what Garrett might bring is, like.

280 00:30:26.690 ⇒ 00:30:28.800 Katherine Bayless: Some of the ideas around, like.

281 00:30:29.490 ⇒ 00:30:33.280 Katherine Bayless: Like, what does this look like as it really scales, like, future.

282 00:30:33.280 ⇒ 00:30:33.640 Uttam Kumaran: Yeah.

283 00:30:33.640 ⇒ 00:30:51.250 Katherine Bayless: You know, to a certain extent. I think that, actually, I mean, future-proofing generally is probably a good skill set to kind of bring to this, because, you know, we are going to get so excited about how fast we can move, like, you know, having some adults in the room that are like, yes, but have you thought about what it’s going to be like to maintain that?

284 00:30:51.250 ⇒ 00:30:59.239 Uttam Kumaran: Yeah, so another piece I’m gonna put here is, like, like, PM… like, tests…

285 00:30:59.470 ⇒ 00:31:05.510 Uttam Kumaran: To run this quarter, to then cascade scalably.

286 00:31:06.330 ⇒ 00:31:09.220 Uttam Kumaran: into… basically into post-CES.

287 00:31:11.900 ⇒ 00:31:15.229 Uttam Kumaran: Into both CES… okay, great.

288 00:31:15.900 ⇒ 00:31:21.600 Uttam Kumaran: Okay, cool. So I can… yeah, I think Garrett’s… I already had him in mind, so he’ll be great.

289 00:31:21.860 ⇒ 00:31:28.610 Uttam Kumaran: So I can adjust the 12 months just to put… so you suggested just putting end of the year?

290 00:31:28.940 ⇒ 00:31:39.689 Katherine Bayless: Yeah, what I would do is, like, put this into two chunks. I would do one that’s the Q2 2026, so that we can demonstrate to Christina, like, we’re not talking about, like, an open-ended retainer, where now you’re buying

291 00:31:40.020 ⇒ 00:31:46.950 Katherine Bayless: martini lunches all week, right? Like, we’re just trying to make this not an administrative burden in between us and delivering value.

292 00:31:47.290 ⇒ 00:31:47.710 Uttam Kumaran: Yeah.

293 00:31:47.710 ⇒ 00:32:00.669 Katherine Bayless: And so, I think if we do the one quarter with this kind of new approach, and she’s like, oh, I get it, then do the next six months as a chunk, and then we’re just done for the year, and then we can do a 27-year-long scope, is kind of my thought.

294 00:32:01.080 ⇒ 00:32:06.380 Uttam Kumaran: Okay, okay. So let’s do that. So what I’m gonna do is I’m gonna put the Q2 2026

295 00:32:06.950 ⇒ 00:32:16.589 Uttam Kumaran: And then, I’m gonna front-load… The, like… Quarterly, plan, admin,

296 00:32:18.060 ⇒ 00:32:32.040 Uttam Kumaran: sort of scope, up front. And then, second thing is, like, I’ll go ahead and just, like, I’m just gonna create a draft of, like, what Q3, Q4, so that in case we need it, or to reference, it’s just gonna be there.

297 00:32:32.300 ⇒ 00:32:38.120 Katherine Bayless: It definitely doesn’t need to be in a final, polished state at this point. Sure. I just wouldn’t want you to lose the work that you’ve put into this, I guess, is more work.

298 00:32:38.120 ⇒ 00:32:52.559 Uttam Kumaran: Sure, sure, sure, sure. No, this is great. I actually think this is, like, I mean, it’s… I just, like, appreciate that you guys are open. Some folks, they’re like, it’s just tough, this is just not… it’s not as traditional, but we have several clients that are on this because of the fact that they move so fast.

299 00:32:52.750 ⇒ 00:32:58.869 Uttam Kumaran: that, like, I was like, hey, we’re gonna take the hit. Like, I will take the hit if, like, things move.

300 00:32:59.060 ⇒ 00:33:17.649 Uttam Kumaran: Because, like, I’m just gonna… we’re gonna deliver, whatever it is. And so, I think this is, like, this is great. So Q3 to Q4 draft. And then the… basically, the way we sort of looked at it is… is we sort of looked at our costs, and it’s not much higher, but I think where… what we kind of did is, like.

301 00:33:18.160 ⇒ 00:33:32.799 Uttam Kumaran: that’s why I sort of had questions about pacing. Like, if you feel like the pacing is good, what you’re gonna see change a little bit is Amber is going to start to drive everything around Coco. I think Garrett is gonna be able to support a bit on the project management side.

302 00:33:32.910 ⇒ 00:33:36.620 Uttam Kumaran: my question to you is, like, how I can…

303 00:33:36.780 ⇒ 00:33:49.869 Uttam Kumaran: elevate it to and support you, like, beyond just managing the team and sort of, like, continuing to… to get us, like, a lot of what I’ve been doing is, like, okay, here’s the first draft of, like, Cortex and, like, getting us

304 00:33:49.930 ⇒ 00:33:58.909 Uttam Kumaran: to solve some of these, like, more hairy, like, the RBAC problem and things like that. But tell me, like, where you… you’d like me to…

305 00:33:59.030 ⇒ 00:34:01.910 Uttam Kumaran: To assist, like, this quarter, or, like, focus more on.

306 00:34:02.190 ⇒ 00:34:10.539 Katherine Bayless: So I think it’s actually a good segue into the other thought I had had, which was, like, I know you kind of have the piece in here about, like, J and AI enablement and stuff like that.

307 00:34:10.540 ⇒ 00:34:24.100 Katherine Bayless: I actually think it might be advantageous to separate those now, and so, like, this scope, this contract is all about, like, the data world, and it obviously involves AI, but it isn’t, like, about the AI stuff.

308 00:34:24.100 ⇒ 00:34:34.159 Katherine Bayless: The reason I say this is partly because it’ll also help with the, like, butter retainer, scary! It’s like, okay, we’re not gonna put the piece that’s the most squishy and unknown yet in here.

309 00:34:34.290 ⇒ 00:34:41.619 Katherine Bayless: I do want it in a different scope, though, where I think we can kind of call out, like, the… and you don’t have to draft that one right away, because I think we’ll talk.

310 00:34:41.620 ⇒ 00:34:41.940 Uttam Kumaran: Sure.

311 00:34:41.949 ⇒ 00:34:50.829 Katherine Bayless: about it too. But, like, you know, figuring out, like, what does Brainforge’s support look like generally? But then I want to have a conversation around, like, how can we leverage you

312 00:34:50.829 ⇒ 00:35:05.759 Katherine Bayless: with our leadership as we sort of navigate these prioritization decisions, right? Like, you know, another adult in the room who can sort of look at a bunch of ideas and say, like, you know, I think these three have a nice overlap and would bring value, and they’d pay dividends down the road for that, and like.

313 00:35:05.759 ⇒ 00:35:09.069 Katherine Bayless: Just some of these sort of, like, can we-should-we type conversations.

314 00:35:09.070 ⇒ 00:35:09.530 Uttam Kumaran: Yeah.

315 00:35:09.530 ⇒ 00:35:21.659 Katherine Bayless: really do want to leverage your expertise in. And then there’s the other piece of it, which is… also kind of benefits from the AI being a different scope, is that Christina’s really trying to figure out, like.

316 00:35:22.400 ⇒ 00:35:47.290 Katherine Bayless: And I mean, totally, I’m like, I don’t think anybody knows the answer yet, but she’s very trying to figure out, like, okay, what’s the difference between working with Catherine’s team on AI stuff and Jay’s team on AI stuff? Yeah. And it being a different scope gives us more space to figure those questions out as we do the work. Because I told her, I was like, I mean, frankly, eventually, it might wind up being my team is more internal-facing, his team’s more external-facing, because that’s kind of traditionally been

317 00:35:47.290 ⇒ 00:35:49.510 Katherine Bayless: a breakdown between data and IT, but, like.

318 00:35:49.510 ⇒ 00:35:51.499 Katherine Bayless: The lines are blurry, and we could go.

319 00:35:51.500 ⇒ 00:35:52.320 Uttam Kumaran: Yeah.

320 00:35:52.440 ⇒ 00:35:57.510 Uttam Kumaran: And also, commonly, we’re seeing teams developing AI teams.

321 00:35:58.160 ⇒ 00:36:09.710 Uttam Kumaran: And so, is that a rebrand of the existing data team? Maybe. Right. Like, that is kind of a natural evolution. But also, yes, like, I think it’s totally fair to separate

322 00:36:09.890 ⇒ 00:36:12.249 Uttam Kumaran: Customer face, because that’s more like product.

323 00:36:12.610 ⇒ 00:36:16.619 Uttam Kumaran: Right? You know, and so that’s a different beast than…

324 00:36:16.790 ⇒ 00:36:23.279 Uttam Kumaran: Just, like, understanding tooling, setting up the layers, that may get leveraged by the product team.

325 00:36:23.470 ⇒ 00:36:32.650 Uttam Kumaran: But, like, so we’re seeing, like, AI teams, we’re still seeing data teams, and then the commercialization of it is a different…

326 00:36:32.780 ⇒ 00:36:36.339 Uttam Kumaran: As a different thing, for sure. So, okay, makes sense.

327 00:36:36.340 ⇒ 00:36:40.340 Katherine Bayless: Interestingly, the commercialization of it is also under me, right?

328 00:36:40.340 ⇒ 00:36:41.320 Uttam Kumaran: Okay, yeah, okay.

329 00:36:41.320 ⇒ 00:37:00.790 Katherine Bayless: like, CES, like, product team is… I mean, and again, it’s starting small purview, and, like, we’ll grow it over time. It’s not like I’m coming in at the top, but, like, that is the idea, is that, like, what is the product of CES in terms of, like, data, AI, digital experience? Like, that’s the thing that I get to try and figure out now, and so, yeah, to your point.

330 00:37:00.790 ⇒ 00:37:07.729 Katherine Bayless: Like, product is different from some of the other stuff, but we have no idea yet what that separation will look like either, so…

331 00:37:07.730 ⇒ 00:37:23.430 Uttam Kumaran: Yeah, and then almost, you know, one way to also think about it is, like, it could just be the infrastructure, like, making sure that this team has access to language models, that they’re provisioned, that, like, we can, like, build on top of them in a structured deployment environment. Like, that’s…

332 00:37:23.610 ⇒ 00:37:25.379 Uttam Kumaran: That is all, like, platform.

333 00:37:25.510 ⇒ 00:37:26.440 Katherine Bayless: Yeah, yeah.

334 00:37:26.440 ⇒ 00:37:27.940 Uttam Kumaran: Or platform work, right?

335 00:37:27.940 ⇒ 00:37:34.879 Katherine Bayless: Right, right. And it’s true, I mean, one of the things that I am very, like, cognizant of is I don’t want more…

336 00:37:34.880 ⇒ 00:37:36.769 Uttam Kumaran: It’s, like, off and stuff like that. Yeah.

337 00:37:36.770 ⇒ 00:37:49.849 Katherine Bayless: Yeah, exactly. Like, I don’t want more security stuff to come my direction, and not because I’m lazy, but because, genuinely, it’s not my area of expertise. I am absolutely competent, and I can do my job as far left as possible, but at the end of.

338 00:37:49.850 ⇒ 00:37:50.410 Uttam Kumaran: Yes.

339 00:37:50.410 ⇒ 00:37:59.010 Katherine Bayless: Realistically, somebody else needs to be the final decision maker where security is concerned, in terms of anything that’s going to be external, or public, or high stakes.

340 00:37:59.010 ⇒ 00:37:59.899 Uttam Kumaran: Yeah, yeah.

341 00:37:59.900 ⇒ 00:38:04.409 Katherine Bayless: So I think that’s a definitely a bright line in the separation eventually.

342 00:38:06.360 ⇒ 00:38:07.150 Uttam Kumaran: Okay.

343 00:38:07.150 ⇒ 00:38:12.449 Katherine Bayless: Unfortunately, Jay’s also not, like, necessarily a big security guy, but it’s okay.

344 00:38:13.120 ⇒ 00:38:24.399 Uttam Kumaran: Okay, so this… this is helpful. So I think I have sort of marching orders on my side. So one is, like, I’ll scope this to just Q2. I’ll tee up at the front, like.

345 00:38:24.490 ⇒ 00:38:36.080 Uttam Kumaran: our conversation about why restructure in this way, why, like, why this is, like, very different than just a traditional flat retainer, like, you’re just gonna get a check.

346 00:38:36.130 ⇒ 00:38:45.520 Uttam Kumaran: And then I will also go ahead, I’m just gonna create the Q3, Q4. It’s just gonna look similar to this, but that way it’s just sitting there, so we can iterate on it.

347 00:38:46.120 ⇒ 00:39:01.560 Uttam Kumaran: I will also… I’m just gonna, like, throw the thoughts on this also into another doc. Okay. That way, like, as I… I’m gonna… I wanna take some time to talk to Jay and learn a bit about this side as well, so that’s where I’ll just start to capture that.

348 00:39:01.880 ⇒ 00:39:04.960 Uttam Kumaran: And then I feel…

349 00:39:05.190 ⇒ 00:39:12.520 Uttam Kumaran: I feel like that’s in a pretty good spot. The other thing is I… I… I am… I was gonna plan on coming next week.

350 00:39:13.140 ⇒ 00:39:13.900 Katherine Bayless: I haven’t booked…

351 00:39:13.900 ⇒ 00:39:19.160 Uttam Kumaran: I haven’t booked it yet, that’s my fault. It’s just been insane. I’m presenting at a conference on Monday.

352 00:39:19.260 ⇒ 00:39:24.330 Uttam Kumaran: But then I was gonna confirm with you what’s a good day for the team.

353 00:39:24.660 ⇒ 00:39:26.519 Uttam Kumaran: Because I would love to…

354 00:39:26.670 ⇒ 00:39:32.580 Uttam Kumaran: to come visit. And then, similarly, I know there’s another… I know there’s some… I know there’s another,

355 00:39:32.810 ⇒ 00:39:34.609 Uttam Kumaran: milestone in May.

356 00:39:35.080 ⇒ 00:39:35.460 Katherine Bayless: Mmm.

357 00:39:35.460 ⇒ 00:39:44.579 Uttam Kumaran: that I think may have been relevant, so I also wanted to bring… see if I can bring Amber, because I think, if I remember correctly, that was more, like, around training for Cortex, or there’s…

358 00:39:45.090 ⇒ 00:39:57.280 Katherine Bayless: We wanted to do something. Yeah, so the May… the May one is on May 12th, that’s when, Jay and I will go back to the CES leadership team with our, sort of, like.

359 00:39:57.400 ⇒ 00:40:11.730 Katherine Bayless: coalesced thoughts around all their AI ideas, and how we’re gonna proceed, and, like, you know, not necessarily saying, like, we’ve chosen the project, but, like, here’s our, you know, guiding policy around how we’re gonna make decisions, here’s where we’re starting, here’s how you’re gonna work with us. I mean, I think even though, like.

360 00:40:11.730 ⇒ 00:40:30.029 Katherine Bayless: Jay has kind of gone a little further in the disposable software direction. Christina still very much wants it to be, like, me and him anytime AI comes up, just because there are these two sides of the house. Yes. And so, like, that’s our chance to go back to that team. We’re also gonna… I’ll do a Snowflake demo to them at that point, so yeah, that’s the May 12th.

361 00:40:30.900 ⇒ 00:40:39.930 Uttam Kumaran: Would you, like… do you want us to come in for that? Like, I would… I think by that point, Amber will be, like, really heads down in everything.

362 00:40:39.930 ⇒ 00:40:48.439 Katherine Bayless: I was gonna say, yeah, like, it might be better to target that date, versus next week, like…

363 00:40:48.650 ⇒ 00:40:52.440 Katherine Bayless: not that you couldn’t come in next week, I mean, if you’re in town, I’d love to at least hang out.

364 00:40:52.440 ⇒ 00:40:52.930 Uttam Kumaran: Yeah.

365 00:40:52.930 ⇒ 00:41:03.620 Katherine Bayless: But I think, like, next week with all the, like, event stuff with CES on the Hill and Digital Patriots, and whatever else going on, like, I think the attention is pretty distracted anyway, and so, like…

366 00:41:04.620 ⇒ 00:41:10.549 Katherine Bayless: since we are already close to it anyway, I think it makes sense to maybe, like, look to May for a more robust.

367 00:41:10.550 ⇒ 00:41:23.990 Uttam Kumaran: Yeah, it was mainly to make the connection with Kai, Kyle, to say hi to you. I actually have to go… I have to be in the East Coast next week, so I was gonna… I was just gonna say hi, so it’s actually, like, I’m gonna be around.

368 00:41:23.990 ⇒ 00:41:39.070 Katherine Bayless: I mean, I would say definitely, like, at least let’s try and hang out, you know? Okay. Even if it’s not, like, an official, like, come to TK kind of thing, but that actually the May timeline is good, too, because Kyle’s wife is due, I think, like, the 24th is the latest date before they will induce

369 00:41:39.070 ⇒ 00:41:46.299 Katherine Bayless: And so he’ll be… I mean, he could disappear any minute now. But by May, he’ll be back, because he’s gonna do two weeks, come back, and then…

370 00:41:46.300 ⇒ 00:41:46.650 Uttam Kumaran: Okay.

371 00:41:46.650 ⇒ 00:41:52.089 Katherine Bayless: two months later in the summer. So May’s actually probably better to catch the whole team, too, anyway, but…

372 00:41:52.090 ⇒ 00:42:01.959 Uttam Kumaran: Okay, so let me… so maybe should we plan for, like, a day or two this May 12th week? Maybe the day before, and that day, I’ll coordinate with my team, and then I’ll just, like.

373 00:42:02.210 ⇒ 00:42:03.809 Uttam Kumaran: Just put a hold on there.

374 00:42:04.360 ⇒ 00:42:11.670 Katherine Bayless: Yeah, okay, so… Monday… So the 11th…

375 00:42:12.080 ⇒ 00:42:21.830 Katherine Bayless: So, the 11th, there will be a lot of people in the office, but we’re also tied up in a culture work… excuse me, culture workshop for the most of the morning, a senior staff meeting in the afternoon.

376 00:42:22.060 ⇒ 00:42:25.760 Katherine Bayless: Tuesday is the meeting with the CES leadership.

377 00:42:26.710 ⇒ 00:42:28.789 Katherine Bayless: Wednesday’s pretty open.

378 00:42:29.210 ⇒ 00:42:32.849 Katherine Bayless: Yeah, and I mean, not that you have to come to the TS leadership meeting anyway.

379 00:42:32.850 ⇒ 00:42:39.879 Uttam Kumaran: No, yeah, I just think maybe 12th and 13, because I think it’d be… if that’s where you’re getting to present, then I think even a debrief

380 00:42:39.990 ⇒ 00:42:49.769 Uttam Kumaran: like, with Amber from my team, and, like, I can even see if Garrett wants to come. I think if we have, like, a renewed focus of, like, okay, where is this heading, could be really nice to do in person.

381 00:42:50.140 ⇒ 00:42:53.990 Katherine Bayless: Yeah, yeah, yeah, actually, that’s a…

382 00:42:53.990 ⇒ 00:43:01.059 Uttam Kumaran: Like, if you leave that meeting, you’re like, okay, like, we have some clarity, we pitched this, this didn’t land, this landed, like, let’s regroup.

383 00:43:01.470 ⇒ 00:43:03.189 Uttam Kumaran: I don’t know, it could be nice.

384 00:43:03.190 ⇒ 00:43:22.990 Katherine Bayless: Yeah, well, and Christina had actually kind of given me a little bit of, like, room to think about, you know, sort of setting up, like, an official sort of retreat type thing, so that, like, she and Jay and I, or whoever else you, right, could kind of have some thought leadership around the idea. So yeah, I think, yeah, 12th… 12th and 13th, or 13th and 14th.

385 00:43:22.990 ⇒ 00:43:26.340 Katherine Bayless: Either one or all three, I think that’s a great window.

386 00:43:26.870 ⇒ 00:43:27.460 Uttam Kumaran: Okay.

387 00:43:28.170 ⇒ 00:43:29.170 Uttam Kumaran: Okay, let’s…

388 00:43:29.170 ⇒ 00:43:37.199 Katherine Bayless: Yeah, and then as far as next week goes, we’re in Monday, Tuesday, Wednesday. If you’re here, you are definitely welcome to stop by. I think happy hour starts at 3, in my opinion, so yeah, just let me know.

389 00:43:37.200 ⇒ 00:43:55.630 Uttam Kumaran: Okay, alright, amazing. I will be there either… I’ll tell you either about Tuesday or Wednesday. It sort of kind of depends on, we have a client meeting. I need to check, my schedule is just all over the place. I am present… I’m giving a talk, and I think it’ll be actually really cool to share with you about AI in the services business.

390 00:43:55.670 ⇒ 00:44:11.319 Uttam Kumaran: And, how we’re… we’re thinking about delivering better outcomes for clients, and, like, how… sharing some client stories with… at this conference, so… should be pretty cool. So I’ll have a lot of thoughts on that when we meet, so…

391 00:44:11.320 ⇒ 00:44:22.219 Katherine Bayless: I’m doing a talk at a conference for, like, it’s the Association for Intelligent Information Management, the week after, so I can, carry forward.

392 00:44:22.220 ⇒ 00:44:22.830 Uttam Kumaran: Okay.

393 00:44:22.830 ⇒ 00:44:23.319 Katherine Bayless: Into that.

394 00:44:23.320 ⇒ 00:44:24.140 Uttam Kumaran: Great.

395 00:44:24.140 ⇒ 00:44:29.609 Katherine Bayless: But yeah, I’m literally, I’m doing a talk about how we’ve decided to, like, you know, just flip everything upside down and inside.

396 00:44:29.610 ⇒ 00:44:30.160 Uttam Kumaran: Yes.

397 00:44:30.160 ⇒ 00:44:30.950 Katherine Bayless: what happens?

398 00:44:31.240 ⇒ 00:44:35.540 Uttam Kumaran: Yeah, I think you guys are, like, probably the first people in your industry to do.

399 00:44:35.980 ⇒ 00:44:47.330 Uttam Kumaran: like this, and the first people, like, even in, I think, industry in general, like, to think about it this way, you’re certainly the first client we’ve had that has been receptive to the, like, maybe we can skip BI.

400 00:44:49.410 ⇒ 00:44:52.980 Uttam Kumaran: You know, and, like, we have a path towards it, you know? So I think that’s, like…

401 00:44:53.320 ⇒ 00:45:03.940 Uttam Kumaran: I think that’s good, that’s an awesome presentation to give, you know. Especially given, like, all the mechanics of CTA, this, like, seasonality around this event, and

402 00:45:04.270 ⇒ 00:45:08.330 Uttam Kumaran: The customer-facing stuff, the internal stuff, yeah, it’s an awesome story.

403 00:45:08.680 ⇒ 00:45:12.970 Katherine Bayless: Yeah, the podcast that I did came out, I think, last week or the week before.

404 00:45:12.970 ⇒ 00:45:15.000 Uttam Kumaran: Oh, wait, I would love to listen to it.

405 00:45:16.540 ⇒ 00:45:22.700 Katherine Bayless: It’s the Reboot IT Podcast. It’s, it’s an association kind of one, but,

406 00:45:22.700 ⇒ 00:45:23.380 Uttam Kumaran: Yeah.

407 00:45:23.510 ⇒ 00:45:27.330 Katherine Bayless: But yeah, I haven’t had the courage to go back and listen to all the things I said.

408 00:45:28.160 ⇒ 00:45:30.470 Uttam Kumaran: I would love to listen to it, that’s amazing.

409 00:45:30.470 ⇒ 00:45:31.580 Katherine Bayless: I’ll find the link.

410 00:45:32.140 ⇒ 00:45:34.850 Uttam Kumaran: Okay, perfect. Yeah, I just found it. Yeah, this is great.

411 00:45:34.850 ⇒ 00:45:37.379 Katherine Bayless: Oh, nice, okay, yeah, yeah, yeah, yeah.

412 00:45:37.940 ⇒ 00:45:38.800 Katherine Bayless: It’s funny.

413 00:45:38.800 ⇒ 00:45:39.359 Uttam Kumaran: Hell yeah.

414 00:45:39.360 ⇒ 00:45:45.220 Katherine Bayless: I consciously tried to not, like, you know, mention too many specific things, partly because, you know, you don’t want to sound like you’re.

415 00:45:45.220 ⇒ 00:45:45.980 Uttam Kumaran: Yeah, yeah, yeah.

416 00:45:45.980 ⇒ 00:45:49.619 Katherine Bayless: But also, because I was like, God only knows, by the time this airs, who knows?

417 00:45:49.620 ⇒ 00:45:56.020 Uttam Kumaran: You know, it’s like, I’m gonna be… I’m gonna… yeah, I’m gonna be a boomer now by the time this comes out.

418 00:45:56.020 ⇒ 00:45:57.150 Katherine Bayless: Exactly.

419 00:45:57.420 ⇒ 00:46:05.280 Katherine Bayless: Right? But yeah, so no, I think… I think it is cool. I mean, I think if we can pull some of this off, like, the case studies we can tell, dude.

420 00:46:05.930 ⇒ 00:46:13.520 Uttam Kumaran: Yeah, I know, that’s why I’m, like, I wanna… I mean, I’ve been procrastinating on just, like, trying to get the Cocoa stuff set up.

421 00:46:13.650 ⇒ 00:46:19.860 Uttam Kumaran: So that I can go have a conversation with Jay and start to show him more about this, like, unified context layer.

422 00:46:19.860 ⇒ 00:46:20.930 Katherine Bayless: Yeah.

423 00:46:21.300 ⇒ 00:46:23.630 Uttam Kumaran: I think I want to move us less into, like.

424 00:46:24.230 ⇒ 00:46:42.159 Uttam Kumaran: whether it’s COCO or the actual, like, the surface, it’s like the thing and the governance around all the knowledge needs to be clear. I think S3 files coming out is a great win for, like, this, so that’s why I think I’m excited to chat with him about that, and…

425 00:46:42.430 ⇒ 00:46:51.599 Uttam Kumaran: And then, yeah, I mean, like, again, I think, for me, I just want to be supportive to you and Jay and Christina on, like, how this whole group wins, so…

426 00:46:51.730 ⇒ 00:46:52.550 Uttam Kumaran: Perfect.

427 00:46:52.950 ⇒ 00:47:01.909 Katherine Bayless: Yeah, no, I think it’s excellent, honestly. And I do keep kind of, like, emphasizing that to people. I’m like, we’re using Snowflake now, but we are building something we can use wherever the market.

428 00:47:01.910 ⇒ 00:47:02.800 Uttam Kumaran: Yeah, yeah.

429 00:47:02.800 ⇒ 00:47:09.269 Katherine Bayless: Like, we will chase the power and the value of our data, not the platform it is on currently.

430 00:47:09.270 ⇒ 00:47:23.580 Uttam Kumaran: Yeah, that’s exactly how I feel like, and there are so many composable open-source ways of achieving that. AWS does a lot, too, but this is sort of proving, like, one narrow, like, end-to-end AI-powered data analysis.

431 00:47:23.690 ⇒ 00:47:40.530 Uttam Kumaran: which is, like, actually a significant technical work, and then you can branch that out into, like, general knowledge work, CRM, you know, and that’s just some of the stuff that I think, if we have a conversation at a high level, I can share about how our sales team is using, like, cursor and skills.

432 00:47:40.530 ⇒ 00:47:41.160 Katherine Bayless: Yeah.

433 00:47:41.160 ⇒ 00:47:57.869 Uttam Kumaran: for example, I built them, like, a person lookup skill, because they’re like, hey, I’m juggling, I’m going to Apollo, Googling LinkedIns, so I built them a skill that, like, they can type in an email, and it runs through a series of analysis to figure out, okay, like, who is this person? We have some waterfall enrichment, and they can now do that

434 00:47:57.990 ⇒ 00:48:10.470 Uttam Kumaran: directly, like, in cursor, or their thing of choice, you know? And then there’s these… it’s very… for mine, it’s very practical. I’m like, you know, you look… oftentimes, you’re looking at people, trying to find their number, trying to find their email to send them.

435 00:48:10.540 ⇒ 00:48:18.339 Uttam Kumaran: a thing, okay, let’s, like, compose that into, like, skills and knowledge, and, like, the surface where you execute the skill, you know, basically.

436 00:48:18.500 ⇒ 00:48:24.099 Katherine Bayless: Exactly. So, like, I was talking to the woman on the membership team the other day, and she was saying that, because

437 00:48:24.100 ⇒ 00:48:48.419 Katherine Bayless: we had a target for membership renewals that we have not yet hit, and I guess even though goals are closed, they’re still chasing it? It’s fine, whatever. But apparently, a number of the companies that had been, you know, non-responsive, right, she’s like, well, now that I have this Sales Intel license, which she literally is like, we’ll have to go to the person’s LinkedIn and use the browser extension to get the sales intel data.

438 00:48:48.420 ⇒ 00:48:49.350 Uttam Kumaran: Yeah, yeah.

439 00:48:49.350 ⇒ 00:49:00.290 Katherine Bayless: she’s like, I realized, like, these companies that hadn’t been responding, like, we just didn’t have the current contacts, and so as soon as I looked up the company in Sales Intel, it gave me, like, you know, the relevant names, and then I reached out to them, and they renewed, and I’m like.

440 00:49:00.420 ⇒ 00:49:01.780 Katherine Bayless: Yeah,

441 00:49:01.780 ⇒ 00:49:03.509 Uttam Kumaran: Wow, let’s go, yeah.

442 00:49:03.510 ⇒ 00:49:11.939 Katherine Bayless: Right? And so I’m like, to your point, if we take that light bulb and we put it in a place that has almost zero friction for them to benefit from that, I mean, forget it, right?

443 00:49:11.940 ⇒ 00:49:23.570 Uttam Kumaran: It’s so obvious for them, yeah, yeah. So another thing is I’m going to enable people in our sales Slack to be able to… I built a little Brainforge Slack agent, and you can trigger the skills now. You can say, like, I want to.

444 00:49:23.570 ⇒ 00:49:23.910 Katherine Bayless: Yeah.

445 00:49:23.910 ⇒ 00:49:36.659 Uttam Kumaran: So you just, like, eliminate the friction, but also everybody in our world is aware of how long these things take. Yeah. So… and I think one thing I was talking to someone yesterday is just sort of removing the decision fatigue, like.

446 00:49:37.050 ⇒ 00:49:53.800 Uttam Kumaran: having to keep making decisions, switch contexts, it’s so taxing just to accomplish one small thing. So, if you can start to… in our side, we’re starting, like, we have, like, templates, playbooks, so the person executing the task.

447 00:49:53.890 ⇒ 00:49:58.279 Uttam Kumaran: Thinks more about the context and, like, checking versus the, like.

448 00:49:58.530 ⇒ 00:50:09.010 Uttam Kumaran: actual step-by-step execution. Yep. You know, whether that’s, like, drafting an email, whether that’s, like, writing a piece of code in the standard way that we always do, or

449 00:50:09.110 ⇒ 00:50:21.120 Uttam Kumaran: like, VR descriptions, even. It’s like, I don’t want people to think about… and what was the past where we have to do run books, people need to, like, read a handbook, or you need to somehow hack together something that, like.

450 00:50:21.460 ⇒ 00:50:33.159 Uttam Kumaran: And it’s brutal. It’s like, nobody is doing that. It’s like… it’s, like, front-of-class type kid who’s like, I did the perfect commit thing. Like, I’m just putting fix, fix, fix, fix.

451 00:50:33.160 ⇒ 00:50:36.140 Katherine Bayless: So, like, how can you…

452 00:50:36.140 ⇒ 00:50:44.889 Uttam Kumaran: expect? Yeah, exactly. How can you… how can you expect it? So now, I think adhering to the standards is really possible, because.

453 00:50:45.410 ⇒ 00:50:52.960 Uttam Kumaran: Because the sell to the knowledge worker is the friction is so low, you know, that, like, it’s not like, oh, it’s gonna take 5…

454 00:50:53.130 ⇒ 00:51:07.760 Uttam Kumaran: 5X longer for me to adhere, then get it done, and then nobody in my org ever follows that anyways. It’s, like, sort of when people are like, oh, like, I worked at a company where, like, I was like, oh, I have to submit, like, out of office. I was like, nobody looks at that, don’t worry about that. I’m like, what?

455 00:51:07.970 ⇒ 00:51:12.060 Katherine Bayless: I mean, it’s so true, though, like, and I, like, even in my little, like.

456 00:51:12.060 ⇒ 00:51:20.980 Uttam Kumaran: You guys are like, it’s the SAP portal, it’s, like, impossible, it doesn’t work. I’m like, okay, just gonna… I guess I’ll just go out of office, whatever. Yeah.

457 00:51:21.060 ⇒ 00:51:32.900 Katherine Bayless: But it’s like, I think, like, if the LLMs and, you know, the coding agents are effectively functioning as natural language compilers, like, I think these scaffolds and skills and, you know, CloudMD files and governance, they’re.

458 00:51:32.900 ⇒ 00:51:33.470 Uttam Kumaran: Yes.

459 00:51:33.470 ⇒ 00:51:44.410 Katherine Bayless: behavioral compilers, right? Because, like, I’m working on my side project, like, the same whirling dervish of chaos I’ve always been from a developer perspective, but I put all these things in so that it cleans up after me, because I know what.

460 00:51:44.410 ⇒ 00:51:45.040 Uttam Kumaran: Yes.

461 00:51:45.040 ⇒ 00:51:55.220 Katherine Bayless: to cause, and I can build things that correct them on the way. And so it’s like, I’ve got this beautiful codebase that my brain would never have built on its own, but I knew which things to hedge against, you know?

462 00:51:55.220 ⇒ 00:52:04.239 Uttam Kumaran: Exactly, but that’s actually what we should be doing. Like, you want the thing to run for a while, try a bunch of things, and then you want to come with the van stuck.

463 00:52:04.620 ⇒ 00:52:21.700 Uttam Kumaran: And then the other things, it’s really the harnessing. So, on our side, I’m trying to help our team develop, okay, you can… you can use a CLI to get into our 1Password easily. We use Railway for a lot of hosting, so you can CLI into Railway. For linear, it’s all MCP, so everybody has that hooked up.

464 00:52:21.700 ⇒ 00:52:33.520 Uttam Kumaran: So then the agents… and then there’s a… in our… in our repo, there’s a lot of instruction on, like, use linear in this situation, use this, and so it will just sort of, like, hit a roadblock, and the reasoning is getting better, where it’ll be like.

465 00:52:33.520 ⇒ 00:52:33.930 Katherine Bayless: Yeah.

466 00:52:33.930 ⇒ 00:52:46.400 Uttam Kumaran: oh yeah, like, I see there’s something related to that. Oh, oh yeah, that’s exactly what I needed. And then it’ll sort of, like, it’ll sort of bumble around, but if… but, like, imagine you have… now, I’d rather have 4 AIs that are bumbling towards the goal.

467 00:52:47.420 ⇒ 00:53:05.190 Uttam Kumaran: And, like, even sitting there and managing, like, one chat window, or, like, God forbid, doing it manually, like, you know, doing, like, looking at the docs. Right. I’m like, I’m like, you look at the docs, and you figure it out.

468 00:53:05.190 ⇒ 00:53:13.530 Katherine Bayless: So I, I came across this repo over the weekend that was, linked intent development, and, you know, just yet another, right, like, experiment in this space.

469 00:53:13.530 ⇒ 00:53:14.169 Uttam Kumaran: Yeah, yeah, yeah.

470 00:53:14.170 ⇒ 00:53:25.419 Katherine Bayless: And so I took it, and I set it up on my side project, and as I was reading its, like, report, I was like, oh, this might actually be the first time I have any idea what’s really happening inside this codebase.

471 00:53:25.700 ⇒ 00:53:31.019 Uttam Kumaran: Yeah, there’s some things where I’m, like, I’m working on some, like, TypeScript stuff, I’m like…

472 00:53:31.020 ⇒ 00:53:31.440 Katherine Bayless: Hmm.

473 00:53:31.440 ⇒ 00:53:33.810 Uttam Kumaran: Hillary said, honestly, I don’t know.

474 00:53:33.810 ⇒ 00:53:34.270 Katherine Bayless: Huh.

475 00:53:34.270 ⇒ 00:53:50.000 Uttam Kumaran: what you need from me, so can you, like, either just dumb it down, or, like, frankly, I just trust your judgment. Yeah. So just, like, you go ahead and make whatever you think is the way out of here. Yeah. But it’s so funny, because I’m like, I don’t know this specific route, like.

476 00:53:50.140 ⇒ 00:53:58.339 Uttam Kumaran: thing. I don’t know anything about, like, node versions. It’s just not… I probably knew a little bit about it at some point for some hacky thing, but…

477 00:53:58.340 ⇒ 00:53:58.840 Katherine Bayless: Honestly.

478 00:53:58.840 ⇒ 00:54:01.619 Uttam Kumaran: If you tell me we need no 22, go ahead and.

479 00:54:01.620 ⇒ 00:54:03.800 Katherine Bayless: Yeah, no, it’s funny, too.

480 00:54:03.800 ⇒ 00:54:06.669 Uttam Kumaran: Frank? Right? Frank?

481 00:54:06.750 ⇒ 00:54:07.790 Katherine Bayless: Like, yeah, I’m like.

482 00:54:07.790 ⇒ 00:54:14.210 Uttam Kumaran: But then when it comes to, like, hey, how should I model this database, or, like, naming conventions, I’m like, yeah, we need underscores, we need this.

483 00:54:15.000 ⇒ 00:54:19.269 Uttam Kumaran: I didn’t want you to create a macro for… like, I’m… so there is, like, where I have, like, specific…

484 00:54:19.540 ⇒ 00:54:33.360 Uttam Kumaran: You know? Another skill, like, you should… I would love to recommend to you, because I’ve been using this like crazy, is this last 30 days skill. I think you would really love this. Basically, I’ve been using this sort of,

485 00:54:33.550 ⇒ 00:54:43.150 Uttam Kumaran: as, like, a research support, when I’m, like, thinking about a product or a tool that I want to build, or a workflow, and I’m, like.

486 00:54:43.470 ⇒ 00:54:47.650 Uttam Kumaran: I’m… instead of Googling, like, how did other people do this.

487 00:54:47.910 ⇒ 00:55:00.819 Uttam Kumaran: Or, like, what is the best practice? I use this, and it basically looks through several, like, open sources for… and it runs a bunch of queries, and then synthesizes, like, a little report.

488 00:55:00.920 ⇒ 00:55:08.359 Uttam Kumaran: So, like, for example, I want… I was trying to implement OpenPanel, which is, like, an open source product analytics, like, alternative to PostHog.

489 00:55:08.490 ⇒ 00:55:23.879 Uttam Kumaran: And I was… but I was… I didn’t know what I… maybe, like, a year ago, I was sort of familiar with Hostog, OpenPanel, a bunch of these things, and I was like, hey, last 30 days, like, what’s the… what’s, like, the, you know, the general industry on, like, open source product analytics?

490 00:55:23.880 ⇒ 00:55:24.410 Katherine Bayless: Yeah.

491 00:55:24.410 ⇒ 00:55:31.160 Uttam Kumaran: And it was able to be, like, postdog is really good, but it’s cloud, open panel, it’s a little bit tougher, but it’s all open source, and, like.

492 00:55:31.530 ⇒ 00:55:37.089 Uttam Kumaran: It was helpful for me to just, like, gather the information about, like, what are people talking about?

493 00:55:37.460 ⇒ 00:55:39.520 Uttam Kumaran: Some of these things are changing, you know?

494 00:55:39.840 ⇒ 00:55:53.539 Katherine Bayless: Right. That’s interesting to think about. I mean, me, I’m immediately gonna start using this, but I also am, like… because, like, I do feel bad for, like, Christina, like, you know, she’s kind of in this, like, spot where she’s gotta sell what I yak about upwards.

495 00:55:53.540 ⇒ 00:55:55.919 Uttam Kumaran: Totally. No, that’s exactly it, yeah.

496 00:55:55.920 ⇒ 00:56:09.539 Katherine Bayless: Right? And so, like, she would probably benefit from some of this too, right? Like, you know, what are some of the things that are going on that are changing our ability? And, like, you know, I, like, I look forward to the day where she comes to me and is like, have you heard of… and I’m like.

497 00:56:09.540 ⇒ 00:56:10.320 Uttam Kumaran: Yes.

498 00:56:10.320 ⇒ 00:56:11.300 Katherine Bayless: Actually, right?

499 00:56:11.300 ⇒ 00:56:22.620 Uttam Kumaran: Yeah, but also that’s… her… her job at that level is to use AI to do that, which is probably a lot of, like, research reports, explained it to me, given my… what I’m trying to do.

500 00:56:22.620 ⇒ 00:56:33.220 Uttam Kumaran: like, what are… I… this is my purview, like, what are some strategy… it’s more strategy-focused than, I think, like, the coding, which is just interesting, but… but, like, again, I think for research.

501 00:56:33.220 ⇒ 00:56:33.580 Katherine Bayless: Yeah.

502 00:56:33.580 ⇒ 00:56:36.789 Uttam Kumaran: I… this is… I think you can easily use this, which is just, like.

503 00:56:37.140 ⇒ 00:56:55.370 Uttam Kumaran: there’s probably some version of, like, look through all the management books, and, like, kind of give me some ideas, I have this challenge, you know, and then one other… one other thing I’m finding helpful is I modified this skill to look at this open alternative site, which is basically, like, open source versions of, like, popular software.

504 00:56:55.370 ⇒ 00:56:57.000 Katherine Bayless: Yeah, yeah, yeah.

505 00:56:57.000 ⇒ 00:57:02.880 Uttam Kumaran: And so, basically, I’m able to take an idea, and it gives me, like, the cloud, the products, it gives me, like.

506 00:57:03.010 ⇒ 00:57:09.639 Uttam Kumaran: Yeah, but there’s also this, like, open source, like, little, like, community-maintained version of a CRM.

507 00:57:09.760 ⇒ 00:57:14.380 Uttam Kumaran: And so, I don’t know, I just found, like, these pairings in the last, like, 30 days.

508 00:57:14.790 ⇒ 00:57:15.179 Katherine Bayless: I can…

509 00:57:15.180 ⇒ 00:57:18.529 Uttam Kumaran: Like, hammering this last 30 days thing, because it’s just been really good.

510 00:57:18.690 ⇒ 00:57:21.589 Katherine Bayless: Yeah, okay. This is cool.

511 00:57:22.120 ⇒ 00:57:40.680 Katherine Bayless: This is really cool. But yeah, like, I think even for her, it could be, like, you know, a component of a process that’s like, you know, what are the weird jargon words that I’m gonna run into in a meeting that I, like, you know, could get a heads up on? Like, you know, could we… could something like this be part of a workflow so that, you know, when somebody throws MCP at her, she’s like, yes, I know what that is, right?

512 00:57:40.680 ⇒ 00:57:47.989 Uttam Kumaran: Yeah, I mean, maybe it’s… maybe it’s as easy as just making, like, a Christina explainer skill, and the skill understands her…

513 00:57:48.070 ⇒ 00:58:01.220 Uttam Kumaran: background, her purview, and understands that these are some of the things that the skill may be used, but also, I think, even in the metaversion, maybe you should just sit next to her and build a skill out.

514 00:58:01.430 ⇒ 00:58:02.350 Katherine Bayless: I have done this.

515 00:58:02.700 ⇒ 00:58:05.920 Uttam Kumaran: Okay, okay, okay, sorry, then I’m just, like, rambling. No, no, I know.

516 00:58:05.920 ⇒ 00:58:26.920 Katherine Bayless: It’s funny, it’s like, I have yet… I’ve done it a couple… tried a couple different angles, and I’ve, like, I’ve yet to find the one that’s sticky, and I’m pretty sure that what I’m gonna do is just, like, one day sit on her computer and, like, set up, something in Cowork, and then also the, like, scheduled task part of it, so that it just, like, alerts her when it has run, rather… because, like, I think that’s the trick, is I haven’t given her something yet that she’s actually gone back to and used.

517 00:58:26.920 ⇒ 00:58:29.369 Katherine Bayless: And so I’m like, okay, I just need this thing to show up in front of your face.

518 00:58:29.720 ⇒ 00:58:35.639 Uttam Kumaran: Yes. Yes. So, then it’s just, like, I think we just make the skills for her and say, use these.

519 00:58:36.320 ⇒ 00:58:40.449 Uttam Kumaran: You know? Or, like, I think you just keep attacking until it gets, like, so easy, and then…

520 00:58:41.470 ⇒ 00:58:53.469 Uttam Kumaran: then when it’s like, well, can you make the skill, like, change it, like, this way? It’s like, okay, yeah, actually you can do that, right? Yep. I think, like, a research report skill, or like a, yeah, Christina research skill, where you can just type in something and it, like.

521 00:58:53.780 ⇒ 00:59:00.280 Uttam Kumaran: uses a mix of these to, like, synthesize and, like, prepare. I think that’s awesome, you know?

522 00:59:00.280 ⇒ 00:59:00.910 Katherine Bayless: Yeah.

523 00:59:01.270 ⇒ 00:59:05.630 Uttam Kumaran: I don’t think I’m using this effectively for, like, more strategy stuff anyways, you know, so…

524 00:59:05.920 ⇒ 00:59:09.879 Katherine Bayless: Yeah, yeah, and I think, like, we can come up with, like, little collections of skills and build out.

525 00:59:09.880 ⇒ 00:59:14.660 Uttam Kumaran: But then she should permeate that to her leadership… to all of her peers and leadership, right? Like…

526 00:59:14.660 ⇒ 00:59:15.280 Katherine Bayless: I think.

527 00:59:15.280 ⇒ 00:59:18.360 Uttam Kumaran: I think that’s, like, that’s a perfect way…

528 00:59:18.490 ⇒ 00:59:28.700 Uttam Kumaran: you know, for it to really tangibly hit those people’s daily workflows, you know? Which I know for their workflows, there’s a lot of meetings, a lot of, like.

529 00:59:28.800 ⇒ 00:59:34.630 Uttam Kumaran: Making decisions with, like, 30% fidelity of the decision, and…

530 00:59:35.020 ⇒ 00:59:39.859 Uttam Kumaran: it’s tough, you know? And so, like, I don’t know how better than supporting, like.

531 00:59:40.270 ⇒ 00:59:45.339 Uttam Kumaran: Either it’s for meeting prep, or it’s for strategy sort of research, or…

532 00:59:45.700 ⇒ 00:59:48.580 Uttam Kumaran: Great, okay. Industry about, like, trade shows.

533 00:59:48.900 ⇒ 01:00:00.190 Katherine Bayless: Yeah, I mean, even though the sort of the use case that, you know, before the, like, coding stuff really took off, that I think, you know, I was having a lot of success with AI was, like, just, like, thought leaders, like, something to talk to, to a certain extent.

534 01:00:00.190 ⇒ 01:00:00.800 Uttam Kumaran: Yes.

535 01:00:00.800 ⇒ 01:00:10.559 Katherine Bayless: like, just a place to, like, workshop and refine ideas. Like, I encouraged the CES leadership to still think about using AI that way as they start to, like.

536 01:00:10.560 ⇒ 01:00:21.569 Katherine Bayless: you know, pursue some of these ideas across their teams, like, use it to help you think about, like, can we do it, should we do it, like, you know, all of those things, like, help it… help you… let it help you brainstorm, basically.

537 01:00:21.570 ⇒ 01:00:22.080 Uttam Kumaran: Yes.

538 01:00:22.080 ⇒ 01:00:30.239 Katherine Bayless: Like, and test your assumptions. Like, if you come up with an idea, and you think it’s great, and yes, it can be done, but, like, you’re assuming these other things will be true.

539 01:00:30.440 ⇒ 01:00:36.179 Katherine Bayless: Right? Like, how risky is that idea? All that kind of stuff. Like, you know, go for a consultant, basically.

540 01:00:36.880 ⇒ 01:00:37.530 Uttam Kumaran: Yes.

541 01:00:39.220 ⇒ 01:00:43.300 Uttam Kumaran: That may be something fun to come to that main meeting with, you know?

542 01:00:43.880 ⇒ 01:00:47.590 Uttam Kumaran: Especially you’re gonna get time on a projector or something?

543 01:00:47.780 ⇒ 01:00:49.550 Katherine Bayless: Oh, no, I’ve got… they’re out at a…

544 01:00:49.550 ⇒ 01:00:50.050 Uttam Kumaran: country break.

545 01:00:50.050 ⇒ 01:00:51.150 Katherine Bayless: Entire hour,

546 01:00:51.150 ⇒ 01:00:52.699 Uttam Kumaran: Oh, then I…

547 01:00:53.680 ⇒ 01:00:57.739 Uttam Kumaran: I… yeah, some show and tell could be really something.

548 01:00:57.740 ⇒ 01:01:05.670 Katherine Bayless: I really want something that I will demo that’ll have, like, like that kind of interviewee sort of component, right?

549 01:01:05.670 ⇒ 01:01:06.580 Uttam Kumaran: Yes.

550 01:01:06.580 ⇒ 01:01:08.319 Katherine Bayless: Yeah. Like, I want to show…

551 01:01:08.320 ⇒ 01:01:11.660 Uttam Kumaran: I see it as just input validation, like, more input validation.

552 01:01:12.350 ⇒ 01:01:16.040 Uttam Kumaran: The AI doesn’t do enough of that, I feel like.

553 01:01:16.200 ⇒ 01:01:16.980 Katherine Bayless: Right? Right?

554 01:01:16.980 ⇒ 01:01:20.100 Uttam Kumaran: Push back! Tell me that I just told you to implement this

555 01:01:20.610 ⇒ 01:01:27.970 Uttam Kumaran: software with, like, literally no instruction. I said, hey, go ahead and implement OpenPanel, I like it.

556 01:01:28.170 ⇒ 01:01:28.819 Katherine Bayless: I’m like…

557 01:01:28.820 ⇒ 01:01:33.340 Uttam Kumaran: You should be like, hey, that’s, like, rude, like, that’s, like, not fair.

558 01:01:35.520 ⇒ 01:01:37.520 Katherine Bayless: Great idea!

559 01:01:37.610 ⇒ 01:01:39.970 Uttam Kumaran: Yeah, great idea, I’m on it! It’s like…

560 01:01:41.030 ⇒ 01:01:44.039 Uttam Kumaran: Really? Like, how far are you gonna get?

561 01:01:44.490 ⇒ 01:01:45.800 Katherine Bayless: Yeah.

562 01:01:46.460 ⇒ 01:01:49.709 Katherine Bayless: I mean, yeah, it just, it really wants to help. It just really wants.

563 01:01:49.710 ⇒ 01:02:00.080 Uttam Kumaran: Yes, yes. I literally described it, I was like, it’s cute, it’s working on it, it’s just, like, bumbling its way to, like, the thing.

564 01:02:00.080 ⇒ 01:02:18.119 Katherine Bayless: That’s how I feel like when I watch Coco work? Like, I literally, I told it, like, I need a bank rec for the finance team out of the AMS data share, and I just, I mean, I watched it spin, but it gave me the thing, right? That I just, like, you know, also was entertained watching the robot figure out on my behalf.

565 01:02:18.510 ⇒ 01:02:31.080 Uttam Kumaran: So another thing that I started to do is, for our cursor, the way I constructed our agent’s MD, it… at the end… at the end of every sort of amount of chat, it’ll say, like, I noticed that you can actually turn this chunk into a skill.

566 01:02:31.480 ⇒ 01:02:31.840 Katherine Bayless: Push.

567 01:02:31.840 ⇒ 01:02:35.489 Uttam Kumaran: It pushes… it pushes the, like, asset creation, or, like.

568 01:02:35.610 ⇒ 01:02:39.789 Uttam Kumaran: Basically, it says, like, now go ahead and, like, improve it for everybody.

569 01:02:39.790 ⇒ 01:02:40.310 Katherine Bayless: Aye.

570 01:02:40.310 ⇒ 01:02:50.850 Uttam Kumaran: of, like, thing. Yeah. It’s something we’re testing internally a lot more, because I want to incentivize people to, like, give back to the next person, you know?

571 01:02:50.850 ⇒ 01:02:51.930 Katherine Bayless: Yeah.

572 01:02:51.930 ⇒ 01:02:52.350 Uttam Kumaran: Yeah.

573 01:02:52.350 ⇒ 01:03:05.669 Katherine Bayless: Actually, I just added this to my little side project, where if I’m chatting with the brain, I suppose, and we come up on something, I can say, like, yeah, make this a skill, make this a thought, make this a part of our workflow, like.

574 01:03:05.670 ⇒ 01:03:06.290 Uttam Kumaran: Yes.

575 01:03:06.290 ⇒ 01:03:07.570 Katherine Bayless: Because I think.

576 01:03:07.570 ⇒ 01:03:08.070 Uttam Kumaran: Yes.

577 01:03:08.070 ⇒ 01:03:09.500 Katherine Bayless: healing processes.

578 01:03:09.630 ⇒ 01:03:10.220 Katherine Bayless: Those are…

579 01:03:10.220 ⇒ 01:03:10.930 Uttam Kumaran: Yes.

580 01:03:10.930 ⇒ 01:03:11.490 Katherine Bayless: That’s the.

581 01:03:11.490 ⇒ 01:03:13.510 Uttam Kumaran: Exactly. Exactly.

582 01:03:14.670 ⇒ 01:03:17.030 Katherine Bayless: yes.

583 01:03:17.560 ⇒ 01:03:24.290 Uttam Kumaran: Okay, I’ll… okay, then next week, I’ll show you a little bit about what some of the stuff we’re internally doing, because I think you’ll be like.

584 01:03:25.070 ⇒ 01:03:32.150 Uttam Kumaran: I feel like some of this stuff, it’s just, it changed… I’m starting to think less about the surfaces and more about these, like.

585 01:03:32.600 ⇒ 01:03:35.579 Uttam Kumaran: Primitives of, like, context.

586 01:03:35.870 ⇒ 01:03:38.039 Uttam Kumaran: making sure that any AI can, like.

587 01:03:38.220 ⇒ 01:03:57.649 Uttam Kumaran: authenticate into other systems, like, I think that could be interesting to share. Yeah. It’s a little… it’s, like, not… it’s a little bit about what I’m presenting, but that’s gonna go over everybody’s head. More talking about, like, how we’ve improved client outcomes, and… and, like, we can have smaller pods, and our engineers are… can actually, like, build relationships, like…

588 01:03:57.700 ⇒ 01:04:09.239 Uttam Kumaran: That’s more of what my friends say, but then I’m gonna… I’m gonna definitely be like, then there’s this context layer, then there’s the memory and the harnesses, and then I’ll be like, talk to me after this if you want to go deeper on that.

589 01:04:10.510 ⇒ 01:04:13.520 Katherine Bayless: I think you’ll have quite a, quite a, quite an audience,

590 01:04:13.520 ⇒ 01:04:15.560 Uttam Kumaran: Yes, yeah.

591 01:04:16.020 ⇒ 01:04:19.070 Uttam Kumaran: It’s funny, it’s like IT… it’s an IT service.

592 01:04:19.230 ⇒ 01:04:27.969 Uttam Kumaran: leadership conference. There’s people in, like, IT services, which is a fun, such a lively crowd, like… Gray suits…

593 01:04:28.890 ⇒ 01:04:29.510 Uttam Kumaran: like.

594 01:04:29.510 ⇒ 01:04:30.020 Katherine Bayless: Huh.

595 01:04:30.020 ⇒ 01:04:36.170 Uttam Kumaran: Yeah, but I’m there, like, I don’t like… I wear… I’m wearing a sweater. I’m gonna be like.

596 01:04:36.380 ⇒ 01:04:51.119 Uttam Kumaran: this is… these are all… none of the slides are gonna have any words on them. I’m gonna just tell a story, and then I’m gonna be like, and here’s, like, part two of the story, here’s all the logos, all these things, would love to chat more.

597 01:04:51.310 ⇒ 01:04:59.319 Katherine Bayless: Yeah, that’s basically what I will wind up doing, is I will, show up, and I will have, that morning decided what I really want to say.

598 01:04:59.320 ⇒ 01:05:00.330 Uttam Kumaran: Yeah, I know, that’s.

599 01:05:00.330 ⇒ 01:05:02.250 Katherine Bayless: Talk about what I do, you know?

600 01:05:02.250 ⇒ 01:05:05.440 Uttam Kumaran: I don’t… I’m not good at preparing for these because I don’t.

601 01:05:05.440 ⇒ 01:05:05.910 Katherine Bayless: Nope.

602 01:05:05.910 ⇒ 01:05:09.869 Uttam Kumaran: present like that. I can’t… I have to tell a story. Yep.

603 01:05:10.090 ⇒ 01:05:18.289 Uttam Kumaran: I like if… and I… it’s honestly for me, like, I just, like, ruminate on the themes in the slides or whatever. It’s, like, not…

604 01:05:18.480 ⇒ 01:05:33.119 Uttam Kumaran: you know, you’re trying to make a connection. Some people are gonna look at that, they’re gonna listen to you, they’re gonna look back at that, they’re gonna… and then most of the time, they’re gonna be thinking about themselves and their job, so you’re just trying to hit them with these, like, pulses of, like, you thought about it this way, thought about it this way, versus, like.

605 01:05:33.580 ⇒ 01:05:39.690 Uttam Kumaran: too long of a, like, a bullet, things like that. I don’t know, but that’s changed over time, I feel like, for me.

606 01:05:40.300 ⇒ 01:05:44.629 Katherine Bayless: Well, yeah, so yeah, totally. I mean, yeah, I used to give people, like, here’s the 3 steps you need, right?

607 01:05:44.630 ⇒ 01:05:45.100 Uttam Kumaran: Oh my god.

608 01:05:45.100 ⇒ 01:05:48.579 Katherine Bayless: questions you might want to ask, right? Like…

609 01:05:48.580 ⇒ 01:05:50.399 Uttam Kumaran: Yeah. Here’s the intuition.

610 01:05:50.400 ⇒ 01:05:51.430 Katherine Bayless: Yeah, great.

611 01:05:51.430 ⇒ 01:06:03.000 Uttam Kumaran: Like, here’s the frustration you’re probably having. Raise your hand. Having that, okay, like, here’s some tools, like, here’s how we kind of think about it, here’s how I got there, given my… my constraints.

612 01:06:03.240 ⇒ 01:06:04.870 Uttam Kumaran: Right?

613 01:06:04.930 ⇒ 01:06:14.970 Uttam Kumaran: And so it’s more of, like, I try to show people that way, and then I’m like, this is one way, I don’t even know whether we’ve, like, hit the nail on the head. It’s working.

614 01:06:15.020 ⇒ 01:06:28.119 Uttam Kumaran: So, like, feel free to take a picture of this, and I don’t know, send it to your Slack, but, like, that’s not really the, like, the goal here is to just try, and here are some things you could try sooner than later.

615 01:06:28.290 ⇒ 01:06:40.360 Katherine Bayless: Right. Yeah, like, I think I’ve gone from trying to, like, give people a specific talent to just, like, sparking curiosity, because it seems like, you know, the better way to do it.

616 01:06:40.360 ⇒ 01:06:41.130 Uttam Kumaran: Yeah.

617 01:06:41.350 ⇒ 01:06:42.120 Uttam Kumaran: Yeah.

618 01:06:42.120 ⇒ 01:06:47.560 Katherine Bayless: But, Claude described me as, productive confusion the other day.

619 01:06:49.180 ⇒ 01:06:53.379 Katherine Bayless: And I was like, I don’t know, I’m gonna take that as a compliment, but I’m not 100% sure.

620 01:06:53.970 ⇒ 01:07:16.639 Uttam Kumaran: I did… I did, like, I… there’s a thing called 16 Assessment, or whatever, where it does the ENTJ, like, personality assessment, and I… my girlfriend and I did it with some other friends, and then it was like, you guys are gonna totally clash, there’s, like, no mesh at all. I was like, this is so funny, because some of my friends really believe in it. I’m kind of like, yeah, but…

621 01:07:17.140 ⇒ 01:07:20.609 Uttam Kumaran: This is such a self-limiting belief, if you start to believe that

622 01:07:20.760 ⇒ 01:07:24.179 Uttam Kumaran: that you’re like, this is… I’m these four letters, and what, like…

623 01:07:25.120 ⇒ 01:07:27.220 Katherine Bayless: It’s all business astrology, right?

624 01:07:27.220 ⇒ 01:07:29.400 Uttam Kumaran: It is, it is.

625 01:07:29.400 ⇒ 01:07:38.090 Katherine Bayless: And the Myers-Briggs in particular was, like, two women who were bored, and pretty smart, I mean, I’ll give them that, but they were, like, bored, and they were like, what if we use this to find boyfriends? I’m like, what?

626 01:07:38.090 ⇒ 01:07:39.440 Uttam Kumaran: Yes, yes.

627 01:07:39.440 ⇒ 01:07:42.190 Katherine Bayless: Like, I think the one woman’s name was Catherine, though, so, you know.

628 01:07:42.190 ⇒ 01:07:48.179 Uttam Kumaran: Okay, alright, smart. Alright, that’s the astrology, like, if the name is Catherine.

629 01:07:48.180 ⇒ 01:07:50.280 Katherine Bayless: Yeah. I mean, right?

630 01:07:50.850 ⇒ 01:07:55.980 Katherine Bayless: It’s like, I feel the same way when people are like, well, I’m an ENTJ as they are, like, when they’re like, I’m a Pisces, and I’m like, what does that mean?

631 01:07:55.980 ⇒ 01:08:13.959 Uttam Kumaran: I usually am like… I’m like, so which one is the good one? Or they’re like, oh, these guys are right. Oh, yeah, yeah, that’s what… I forgot, that’s exactly, like, the one I am. And then I’ll be with friends who are like, dude, that’s not… that’s not right. I’m like, I’m just trying to have fun in this conversation. I just…

632 01:08:14.080 ⇒ 01:08:19.279 Uttam Kumaran: I’m like, oh my god, exactly. I totally saw it, could see that for you. I’m like.

633 01:08:19.609 ⇒ 01:08:22.219 Uttam Kumaran: Right? Yeah. Great.

634 01:08:22.229 ⇒ 01:08:23.869 Katherine Bayless: I knew you were a Pisces.

635 01:08:23.870 ⇒ 01:08:24.810 Uttam Kumaran: Yeah.

636 01:08:24.819 ⇒ 01:08:27.589 Katherine Bayless: Right? Yeah. No. It’s very silly.

637 01:08:27.590 ⇒ 01:08:32.009 Uttam Kumaran: I’m too much of an engineer, I can’t, like, how… I don’t know, don’t think like that, so…

638 01:08:32.010 ⇒ 01:08:36.640 Katherine Bayless: Right, yeah, no, I’m a systems thinker, at, like, to a pathological level, like.

639 01:08:36.640 ⇒ 01:08:37.420 Uttam Kumaran: Yes.

640 01:08:37.420 ⇒ 01:08:40.589 Katherine Bayless: like, part of that is, like, being the chameleon, right? I’m like, whatever the.

641 01:08:40.590 ⇒ 01:08:40.960 Uttam Kumaran: Yeah.

642 01:08:40.960 ⇒ 01:08:44.310 Katherine Bayless: I will find it, I will figure it out, and then I will navigate it, and that means.

643 01:08:44.319 ⇒ 01:08:48.059 Uttam Kumaran: Yeah, I’ll be like, yeah, I wonder… I was like, I woke up the other day, I was like.

644 01:08:48.209 ⇒ 01:08:58.719 Uttam Kumaran: I, like, when you turn the water in the house on, like, does it heat automatically, or how does the reservoir work? And, like… my grandma’s like, why are you, like, what do you think about? I’m like.

645 01:08:58.979 ⇒ 01:09:08.539 Uttam Kumaran: it just… at this house, it got hot really fast. At the other house, it didn’t. I’m, like, trying to think… I’m like, how does the mechanics work? I gotta watch some YouTube videos today. That’s how I’m thinking about it.

646 01:09:08.729 ⇒ 01:09:10.429 Uttam Kumaran: yes.

647 01:09:10.430 ⇒ 01:09:14.059 Katherine Bayless: Yeah, the problem is, brains like ours think about everything that way.

648 01:09:14.060 ⇒ 01:09:17.170 Uttam Kumaran: Yeah, it’s too much. It’s too much.

649 01:09:17.170 ⇒ 01:09:31.009 Katherine Bayless: Like, honestly, the early days after it was, like, the initial, like, ChatGPT magic trick excitement, like, once it started to be really kind of sticky, like, I almost burned myself out just, like, suddenly being able to ask every stupid question that came into my.

650 01:09:31.010 ⇒ 01:09:31.590 Uttam Kumaran: Yes.

651 01:09:31.590 ⇒ 01:09:32.240 Katherine Bayless: And eventually, I was.

652 01:09:32.240 ⇒ 01:09:32.680 Uttam Kumaran: Yes.

653 01:09:32.680 ⇒ 01:09:34.629 Katherine Bayless: Actually, I don’t think this is a good thing.

654 01:09:34.630 ⇒ 01:09:37.450 Uttam Kumaran: Actually, I can do things on my own.

655 01:09:37.500 ⇒ 01:09:41.050 Katherine Bayless: Well, or just, like, some things can remain a mystery. I don’t need every answer.

656 01:09:41.050 ⇒ 01:09:46.709 Uttam Kumaran: Exactly, yeah, yeah, actually, that’s more of the magic. Yeah, like, even for gardening and stuff, I’m like.

657 01:09:47.420 ⇒ 01:09:52.050 Uttam Kumaran: I think you need to just develop a feel, and if… if it dies, then…

658 01:09:52.180 ⇒ 01:10:04.330 Uttam Kumaran: You machine learn, you know? Then we machine learn. Like, I don’t need to, like, get the spacing for… to get the ruler for the… just put the thing in the dirt. This is supposed to be natural, you know?

659 01:10:04.510 ⇒ 01:10:10.739 Uttam Kumaran: I’m like, how do I maximize my arugula, like, productions? Like…

660 01:10:11.090 ⇒ 01:10:12.369 Katherine Bayless: So bad.

661 01:10:12.370 ⇒ 01:10:13.630 Uttam Kumaran: Every time.

662 01:10:13.630 ⇒ 01:10:14.920 Katherine Bayless: Every time, yeah.

663 01:10:14.920 ⇒ 01:10:15.450 Uttam Kumaran: Yeah.

664 01:10:15.450 ⇒ 01:10:27.680 Katherine Bayless: Yeah, or I probably saw that somewhere on my phone somewhere. Like, a screen grab of a Reddit post where somebody wrote in, like, you know, like, a life pro tip, plant a tree so that in 20 years, your kids will have shade or something.

665 01:10:27.680 ⇒ 01:10:28.320 Uttam Kumaran: Yeah, yeah, yeah.

666 01:10:28.320 ⇒ 01:10:34.080 Katherine Bayless: I’ll tell you how my brain works, right? Like, I will then decide, okay, I need to learn everything about trees, and then I go.

667 01:10:34.500 ⇒ 01:10:37.190 Katherine Bayless: a rabbit hole, and then the, like, end of it is, like, and then you know…

668 01:10:37.190 ⇒ 01:10:52.820 Uttam Kumaran: Tell me about shade production, what grows here? You know, like, I have a really hard time with that, too. But that… I’ve turned into my job, which has been, in one aspect, really productive. But that tends to not lend itself well to just

669 01:10:53.220 ⇒ 01:11:01.299 Uttam Kumaran: So, that’s for me, like, I real… if I’m just, like, sitting on the couch, I’m just, like, brain dead. I cannot do anything, or I’m, like.

670 01:11:01.430 ⇒ 01:11:05.640 Uttam Kumaran: 100% on. I… there’s not much of a middle ground, like…

671 01:11:05.640 ⇒ 01:11:06.900 Katherine Bayless: Yeah. For me.

672 01:11:06.900 ⇒ 01:11:24.180 Uttam Kumaran: At all. Like, I need… I’m like, I enjoy sitting, having a glass of wine in the courtyard at a bar, and I’m like, just watch… not doing anything. And at work, I’m like, I have, like, 4 different open code agents running, like, it’s insane.

673 01:11:24.470 ⇒ 01:11:27.850 Katherine Bayless: I mean, at home, I’m absolutely that girl with both laptops up, right?

674 01:11:27.850 ⇒ 01:11:28.410 Uttam Kumaran: Yeah.

675 01:11:28.410 ⇒ 01:11:31.029 Katherine Bayless: Little this, little of that, little this, little of that, but then…

676 01:11:31.030 ⇒ 01:11:36.249 Uttam Kumaran: I’m like, oh, I didn’t… I didn’t eat, I wouldn’t, like, when I drink water, like, where am I?

677 01:11:36.250 ⇒ 01:11:37.639 Katherine Bayless: it’s midnight.

678 01:11:37.640 ⇒ 01:11:38.420 Uttam Kumaran: Right, yes.

679 01:11:38.420 ⇒ 01:11:45.110 Katherine Bayless: Right. Exactly, oh my god. But then it’s like, I sit on the metro, and I don’t even, like, listen to music, I just, like, stare into space, right?

680 01:11:45.110 ⇒ 01:11:45.620 Uttam Kumaran: Good night.

681 01:11:45.620 ⇒ 01:11:48.840 Katherine Bayless: Nice. Just like, that’s my chance to just veg out.

682 01:11:48.840 ⇒ 01:11:50.170 Uttam Kumaran: Recharge, yeah.

683 01:11:50.470 ⇒ 01:11:56.290 Katherine Bayless: Although there are mornings where I’m like, but I could… I could hotspot my phone. I could get started on some things.

684 01:11:56.450 ⇒ 01:12:05.609 Uttam Kumaran: Yeah, yeah, or I’ll, like, I used to listen to a lot of work podcasts, and now I can’t. It’s just so much work. Yeah. So instead, I listen to just comedy podcasts.

685 01:12:06.220 ⇒ 01:12:14.070 Uttam Kumaran: and it’s just… it’s just people talking whatever, and I’m like, this is just nice, around friends, just chatting.

686 01:12:14.730 ⇒ 01:12:16.790 Katherine Bayless: It’s same, same, like I…

687 01:12:16.790 ⇒ 01:12:32.510 Uttam Kumaran: Like, something totally left field, nothing to do. But then sometimes they’ll talk about AI, I’m like, no, like, they’re like, oh yeah, I was using Claude, it’s all cloud code. I’m like, no, like, please do not let this weave into this podcast.

688 01:12:32.510 ⇒ 01:12:35.340 Katherine Bayless: It’s, like, contagion spreading. You’re, like, trying to.

689 01:12:35.340 ⇒ 01:12:35.800 Uttam Kumaran: Yeah.

690 01:12:35.800 ⇒ 01:12:37.269 Katherine Bayless: Yeah,

691 01:12:37.270 ⇒ 01:12:37.790 Uttam Kumaran: S.

692 01:12:37.960 ⇒ 01:12:54.230 Katherine Bayless: No, I know, it’s funny, like, I… it is something I think a lot about, like, what are the, like, boom and bust cycles? How do we, like, still make mental space for, like, rest and reflection, especially with so many of our thoughts kind of now parked externally, not even.

693 01:12:54.230 ⇒ 01:12:54.820 Uttam Kumaran: Yeah.

694 01:12:54.820 ⇒ 01:12:58.789 Katherine Bayless: Available for the normal processes to run over them and, you know, while.

695 01:12:58.790 ⇒ 01:13:02.160 Uttam Kumaran: But I still think it’s people… people interact with people, like, in one…

696 01:13:02.750 ⇒ 01:13:17.949 Uttam Kumaran: glass half full is, you would say, we’re the limiting factor. Glass half full is, like, okay, you still need to convince and share and teach, and so that’s the aspect I think a lot about, is like.

697 01:13:18.110 ⇒ 01:13:31.329 Uttam Kumaran: nothing hap… even if you have 10 things running or whatever, nothing happens unless you improve the leverage for your team, your company. Who cares? It doesn’t matter unless everybody picks it up. So that’s what I’ve kind of tried to think more about, is, like.

698 01:13:31.480 ⇒ 01:13:38.620 Uttam Kumaran: I try to run into the future, and then I, like, come back, and then I do another run to the future, then I come back.

699 01:13:38.840 ⇒ 01:13:44.179 Uttam Kumaran: And, like, I think that’s what’s been helping me a lot, because there’s still people in my company who are, like.

700 01:13:44.740 ⇒ 01:13:49.349 Uttam Kumaran: they just wrote the skill… the first skill, like, a few weeks ago, and I’m like, okay, like…

701 01:13:49.910 ⇒ 01:13:55.049 Uttam Kumaran: I was here, like, 6 months ago, but okay, we’re good, we’re making progress, like, we’re getting there, you know?

702 01:13:55.280 ⇒ 01:13:59.909 Katherine Bayless: Yeah, I mean, it’s, you know, it’s the William Gibson quote, right? The future is here, it’s just not evenly distributed.

703 01:13:59.910 ⇒ 01:14:02.000 Uttam Kumaran: Distributed, yeah. Yeah.

704 01:14:02.000 ⇒ 01:14:09.950 Katherine Bayless: Yeah, it’s funny, it’s actually something I’ve been trying to work on with Kyle, and then this other guy, Chris, who’s on the market research team from, like, a mentorship sort of perspective.

705 01:14:09.950 ⇒ 01:14:10.600 Uttam Kumaran: Yeah.

706 01:14:10.600 ⇒ 01:14:11.379 Katherine Bayless: is, like.

707 01:14:12.050 ⇒ 01:14:23.330 Katherine Bayless: they already are the type of person who tends to have the right answer, not just think they have, they probably have the right answer for something. And now AI, like, accelerates that, and it, like, you know, triples down on, like.

708 01:14:23.330 ⇒ 01:14:34.410 Uttam Kumaran: And if they’re commonly frustrated when people don’t, it could just… it’ll just exacerbate that. Because I don’t… I’m not like that. I’m like, just tell me where you are so I can show you how to jump, like…

709 01:14:34.540 ⇒ 01:14:44.709 Uttam Kumaran: That’s it. So, wherever anyone is. But I also have a lot of friends who are like that. They’re like, oh, I figured it out, like, and I… it’s like… it’s like pushing too much pressure on them, you know?

710 01:14:44.710 ⇒ 01:15:00.299 Katherine Bayless: Yeah! Well, exactly, so it’s like, we had this, like, Chris wanted to do some work with, like, social media, like, web scraping type data, and the marketing team has a subscription to TalkWalker to do social media monitoring, and so, like, he was like, gimme, I want.

711 01:15:00.300 ⇒ 01:15:11.789 Katherine Bayless: Right? And they were like, whoa, we already do this, what are you at? Like, what’s, you know, hang on, what’s happening here, right? And then, like, the conversation was kind of, like, DOA for a while, until…

712 01:15:12.020 ⇒ 01:15:15.579 Katherine Bayless: I kind of just, you know, gently planted some little seeds around, and then.

713 01:15:15.580 ⇒ 01:15:15.980 Uttam Kumaran: Yeah.

714 01:15:15.990 ⇒ 01:15:23.059 Katherine Bayless: marketing came to us and was like, so there’s a Fivetran connector for TalkWalker, can we bring that data into Snowflake? And I’m like.

715 01:15:23.930 ⇒ 01:15:24.600 Uttam Kumaran: Yes.

716 01:15:25.560 ⇒ 01:15:29.639 Uttam Kumaran: But that’s… but the thing is, you should show Chris, like, okay, this is, like…

717 01:15:29.800 ⇒ 01:15:44.469 Uttam Kumaran: Right. How you cement the seeds, but also, everything can be framed as a win-win. Everything is a win-win. You just have to think about it that way. But that is the engineering part. Like, I think about that as an engineering problem. Absolutely. I’m like, yo, you use Cloud Code, you do…

718 01:15:44.590 ⇒ 01:16:04.379 Uttam Kumaran: Yeah, easy. Like, that’s… that’s the easy part about all this. It’s like, now, like, you have to convince these folks that you’re to be trusted, and that you are… have their best interest, you then deliver on that. Yep. And, like, I think… but I don’t know, for some people, frame… for me, framing that as a problem helped me, because I’m like, okay, let me learn how to solve that.

719 01:16:04.740 ⇒ 01:16:13.329 Uttam Kumaran: problem, you know? Instead of, like, oh, this… you have to play politics, or people are in the way. Instead, it’s like, no, these are… this is just how…

720 01:16:13.460 ⇒ 01:16:17.059 Uttam Kumaran: Your objective is here. You cannot achieve that without

721 01:16:17.430 ⇒ 01:16:26.290 Uttam Kumaran: convincing these folks, and befriending these folks, and… and delivering wins for these folks. So, that should be freeing. I’ve given you the answer. Right?

722 01:16:26.290 ⇒ 01:16:33.449 Katherine Bayless: Honestly, I mean, that’s exactly… I’m trying… trying to get them there. There’s a lot of folks here who are more in the camp of, like, well, just tell them to do it.

723 01:16:33.450 ⇒ 01:16:43.140 Uttam Kumaran: Look, if you’ve been burned a lot, I get it, like, if you’ve been burned, or someone did that, if… a lot of things, baggage from, like, that’s just how we work around here, and so it’s like, okay, like…

724 01:16:43.320 ⇒ 01:16:47.660 Uttam Kumaran: You’re now coming with a solution that’s so much better that you have leverage.

725 01:16:47.810 ⇒ 01:16:53.460 Katherine Bayless: That’s exactly, exactly right. I’m like, look, we’re not coming at them with Power BI 20 years too late.

726 01:16:53.460 ⇒ 01:16:53.980 Uttam Kumaran: Yeah.

727 01:16:53.980 ⇒ 01:17:01.939 Katherine Bayless: with the future, and that’s different and scary, but if we give them the space to figure out what part of the future they are willing to engage with.

728 01:17:01.940 ⇒ 01:17:02.430 Uttam Kumaran: Yeah.

729 01:17:02.430 ⇒ 01:17:03.360 Katherine Bayless: Then we get to.

730 01:17:03.360 ⇒ 01:17:05.669 Uttam Kumaran: And it’s just a couple meetings of listening, it’s a.

731 01:17:05.670 ⇒ 01:17:06.080 Katherine Bayless: meeting.

732 01:17:06.080 ⇒ 01:17:15.250 Uttam Kumaran: and asking questions, finding out, like, what is the… what is the hesitation? And you’ll find that they’ll just… people will just tell you exactly, like, what the hesitation is.

733 01:17:15.730 ⇒ 01:17:23.960 Katherine Bayless: well, like, I told Chris, because he used to work in consulting, I was like, you know, think about what every consultant says when they start an engagement. How do I get you promoted? I was like.

734 01:17:23.960 ⇒ 01:17:24.400 Uttam Kumaran: Yes.

735 01:17:24.400 ⇒ 01:17:29.729 Katherine Bayless: That’s what you need to be thinking about in every conversation, regardless of whether or not you actually want that person promoted.

736 01:17:29.730 ⇒ 01:17:31.110 Uttam Kumaran: Yeah, yeah.

737 01:17:31.110 ⇒ 01:17:31.870 Katherine Bayless: Right? Yeah.

738 01:17:31.870 ⇒ 01:17:32.480 Uttam Kumaran: Yeah.

739 01:17:32.480 ⇒ 01:17:35.350 Katherine Bayless: I help you in a way that builds this network?

740 01:17:35.610 ⇒ 01:17:39.909 Uttam Kumaran: Yeah. That’s great, that’s good advice for them to hear, yeah, that’s great.

741 01:17:40.280 ⇒ 01:17:46.369 Katherine Bayless: I think the most adorable part is how much they grumble when I say it, but then, like, two days later, they’ll come back and be like, I was realizing that.

742 01:17:46.370 ⇒ 01:17:52.390 Uttam Kumaran: Because if they’re truly objective, like, people that want to win, they will realize that that is

743 01:17:52.710 ⇒ 01:17:56.669 Uttam Kumaran: This is part of the puzzle. If they’re not, then they’re not serious.

744 01:17:56.670 ⇒ 01:17:57.750 Katherine Bayless: Yeah. You know? Yep.

745 01:17:57.750 ⇒ 01:18:12.669 Uttam Kumaran: then they’re just engineering. Like, that’s… that’s why I tell people, if you’re serious, and you want to win, and you want to increase your leverage and your impact, this is the only way. But treat it like another puzzle. So for more engineering minds, it helps them, because that’s how it helps me. I’m not…

746 01:18:12.770 ⇒ 01:18:17.989 Uttam Kumaran: If I didn’t have to do it, I wouldn’t, but this is what success… this is how we succeed. You know, it’s very easy.

747 01:18:18.170 ⇒ 01:18:22.229 Katherine Bayless: This is how you parlay a French degree into running tech for CES.

748 01:18:22.230 ⇒ 01:18:24.980 Uttam Kumaran: Yes, yeah, you just win for people, yeah.

749 01:18:24.980 ⇒ 01:18:28.280 Katherine Bayless: Yeah, yeah, so yeah, same page.

750 01:18:28.470 ⇒ 01:18:30.040 Uttam Kumaran: Okay, cool, cool.

751 01:18:30.040 ⇒ 01:18:32.309 Katherine Bayless: Oh my gosh, I don’t even get to have a drink next week.

752 01:18:32.310 ⇒ 01:18:45.349 Uttam Kumaran: Yeah, yeah, definitely. So let me get this, let me, let me, like, fix some of this, and then I’ll tell you when I’m there next week, and then I will also, just sometime this week, go ahead and plan with Amber and Garrett about… about May, so…

753 01:18:45.350 ⇒ 01:18:46.580 Katherine Bayless: Okay, yeah, yeah, yeah, yeah.

754 01:18:46.840 ⇒ 01:18:53.459 Uttam Kumaran: And I know we missed this Cocoa meeting, but I’m excited to go check out if they ended up figuring this stuff out, so…

755 01:18:53.460 ⇒ 01:19:12.209 Katherine Bayless: Yeah, I mean, honestly, I think the biggest question I had for the meeting was just, like, I think… actually, I will probably use the last 30 days, repo to answer it, because I was just gonna ask you guys, like, okay, I know we’re kind of going in this one direction, but, like, are we getting good coverage on the various and sundry AI things Snowflake does? But, that’s me.

756 01:19:12.210 ⇒ 01:19:19.449 Uttam Kumaran: I told Amber the next thing we’re gonna look at is agents, but I told her nothing we’re gonna do in agents is gonna succeed without, like.

757 01:19:19.600 ⇒ 01:19:23.400 Uttam Kumaran: the semantic view layer, so I said, like, I want you to nail that, I’ll go…

758 01:19:23.530 ⇒ 01:19:26.369 Uttam Kumaran: I’ll go a little bit into the future and, like, test some agents.

759 01:19:26.370 ⇒ 01:19:27.170 Katherine Bayless: And then…

760 01:19:27.170 ⇒ 01:19:31.840 Uttam Kumaran: she’ll knock that out. Mainly, I wanted her to just connect with Kyle and… and Kai, and, like.

761 01:19:32.140 ⇒ 01:19:35.229 Uttam Kumaran: Start to connect the dots, because she’ll start to sprint, so…

762 01:19:35.320 ⇒ 01:19:47.369 Katherine Bayless: Yeah, well, and I think also you’re… like, you and I are in the same space where it’s like, agents, yes, no, maybe, sure, fine, whatever, worth looking into, but, like, probably the thing that we really will benefit going forward is the skills, and so it’s like…

763 01:19:47.370 ⇒ 01:19:48.050 Uttam Kumaran: Yeah.

764 01:19:48.050 ⇒ 01:19:49.440 Katherine Bayless: We maybe dabble with the agency.

765 01:19:49.440 ⇒ 01:19:53.439 Uttam Kumaran: I don’t want to offload all of the, like, execution until…

766 01:19:53.440 ⇒ 01:19:53.870 Katherine Bayless: Right.

767 01:19:53.870 ⇒ 01:19:59.909 Uttam Kumaran: we have understood the workflows really well, because then those agents will just execute the workflow.

768 01:20:00.430 ⇒ 01:20:05.320 Uttam Kumaran: But again, part of this is, like, if it helps for me to have, like, an agent show, like.

769 01:20:05.510 ⇒ 01:20:06.930 Uttam Kumaran: It’s reasoning…

770 01:20:07.140 ⇒ 01:20:14.479 Uttam Kumaran: But I still, like, nothing happens without the semantic view, so we’ll just push on whatever pedals, like, we have in front of us.

771 01:20:14.730 ⇒ 01:20:20.469 Uttam Kumaran: You know, to see what’s working. Because also, maybe we can start using ages on our team if we’re, like, sophisticated enough.

772 01:20:21.510 ⇒ 01:20:34.920 Katherine Bayless: Yeah. You know? Yeah, I mean, I think it’s always kind of my stance, too, is like, let us be the guinea pigs wherever possible, right? And then, you know, when it comes time to roll it out further, we can say, like, well, yeah, we started, and we ran into some quirks, but, you know, here’s what we’ll do, and yeah, yeah.

773 01:20:35.160 ⇒ 01:20:35.600 Uttam Kumaran: Yeah.

774 01:20:35.600 ⇒ 01:20:36.220 Katherine Bayless: Yeah.

775 01:20:36.220 ⇒ 01:20:36.740 Uttam Kumaran: Okay.

776 01:20:36.980 ⇒ 01:20:38.409 Katherine Bayless: But yeah, awesome.

777 01:20:38.630 ⇒ 01:20:45.149 Uttam Kumaran: Okay, great chat, this was great. And I’m gonna listen to this podcast later, so I’m excited.

778 01:20:45.220 ⇒ 01:20:48.869 Katherine Bayless: Let me know what you think. All feedback is welcome.

779 01:20:48.870 ⇒ 01:20:50.599 Uttam Kumaran: Absolutely. Okay, okay, perfect.

780 01:20:51.020 ⇒ 01:20:51.840 Katherine Bayless: Alright, see you.

781 01:20:51.840 ⇒ 01:20:53.630 Uttam Kumaran: Okay, thank you, bye.