Meeting Title: Brainforge: Uttam <> Elizabeth Date: 2025-10-16 Meeting participants: Elizabeth Gilbert, Uttam Kumaran
WEBVTT
1 00:00:59.170 ⇒ 00:01:00.559 Uttam Kumaran: Hey, how are you?
2 00:01:01.230 ⇒ 00:01:02.869 Elizabeth Gilbert: Hello, I’m doing well, how are you?
3 00:01:02.870 ⇒ 00:01:04.979 Uttam Kumaran: Good, nice to meet you, thanks for taking the time.
4 00:01:04.989 ⇒ 00:01:06.589 Elizabeth Gilbert: Me too, as well, yeah.
5 00:01:06.920 ⇒ 00:01:08.990 Uttam Kumaran: Where are you, in the world right now?
6 00:01:09.750 ⇒ 00:01:13.060 Elizabeth Gilbert: I’m in Cincinnati, Ohio, at a co-working space I go to.
7 00:01:13.190 ⇒ 00:01:18.220 Uttam Kumaran: Nice, okay, I’m at home. This is my co-working space. I’m in Austin.
8 00:01:18.930 ⇒ 00:01:19.620 Elizabeth Gilbert: Fairness.
9 00:01:19.810 ⇒ 00:01:35.840 Uttam Kumaran: Yeah. Well, awesome. Yeah, I appreciate you taking the time. Maybe I’ll give you a little bit of background. So, I run a company called Brainforge. We’re a data analytics and AI consultancy. My background is in data engineering. I worked as a data engineer
10 00:01:36.260 ⇒ 00:01:41.920 Uttam Kumaran: For a while, and then started working, in leading data teams, and then led product at a data startup.
11 00:01:42.040 ⇒ 00:01:55.870 Uttam Kumaran: And then started this company roughly 2 years ago, and so we’ve slowly been growing. We’re roughly, like, 15 people now. We’ve helped almost, like, 30 or so companies.
12 00:01:56.260 ⇒ 00:02:07.100 Uttam Kumaran: data or AI-related, projects. But, you know, we do a lot of stuff in, like, the classic modern data stack, so setting up warehouses, ETLs.
13 00:02:07.930 ⇒ 00:02:09.440 Uttam Kumaran: doing a lot of modeling in dbt.
14 00:02:09.449 ⇒ 00:02:09.809 Elizabeth Gilbert: God.
15 00:02:09.810 ⇒ 00:02:15.640 Uttam Kumaran: And then, you know, we do a lot of sales, finance, marketing-related, analysis.
16 00:02:15.960 ⇒ 00:02:24.860 Uttam Kumaran: And so, we typically work with, like, private businesses, so anywhere from, like, $20 million to, like, a couple hundred million, I would say.
17 00:02:24.860 ⇒ 00:02:25.450 Elizabeth Gilbert: We…
18 00:02:25.450 ⇒ 00:02:29.849 Uttam Kumaran: Typically come in as part of their data function, or in some cases, we run the entire.
19 00:02:29.850 ⇒ 00:02:30.230 Elizabeth Gilbert: That’d be good.
20 00:02:30.230 ⇒ 00:02:39.420 Uttam Kumaran: data function. And then my background is working a lot with executives, so we do help a lot on, like, procuring the right tools, setting up data teams, and then, of course.
21 00:02:40.010 ⇒ 00:02:40.580 Uttam Kumaran: just.
22 00:02:40.580 ⇒ 00:02:41.430 Elizabeth Gilbert: Like, the… the…
23 00:02:41.430 ⇒ 00:02:44.650 Uttam Kumaran: day-to-day. And yeah, we’re, we’re, like.
24 00:02:44.750 ⇒ 00:02:59.270 Uttam Kumaran: you know, always in the lookout for, you know, great data people. I don’t think we hire, like, particularly fast, but, sort of looking for people that have worked in sort of similar environments. Kind of the joy of, like, our work is we get to walk into
25 00:02:59.370 ⇒ 00:03:23.460 Uttam Kumaran: many different types of companies, but a lot of the classic principles of data still apply. Like, we’re helping people make more decisions and make more informed decisions, but we also still help a lot with, like, just picking the right tools, and how do you, like, conduct an analysis, or how do you set up an A-B testing sort of framework, and things like that. So, yeah, that’s a little bit about us. I mean, we’re, again, we’re…
26 00:03:23.910 ⇒ 00:03:30.260 Uttam Kumaran: We’re sort of interested in seeing, like, if there are folks in the market that are interested in doing this type of stuff with us, and how we can partner.
27 00:03:30.790 ⇒ 00:03:32.520 Uttam Kumaran: So yeah, that’s a little bit about us.
28 00:03:33.180 ⇒ 00:03:49.950 Elizabeth Gilbert: Awesome, yeah, I appreciate the overview. One question I have is, it sounds like you do a lot of, like, data engineering in, warehousing ETL, kind of dbt, specifically. Do you also work in, like, clickstream and interaction data as well?
29 00:03:50.400 ⇒ 00:04:09.700 Uttam Kumaran: Yeah, so we, we… so in companies where it is really… either has a… they have a digital product, or they have a clear, like, funnel-based process, yeah, so we typically use… we’re partners with Amplitude and Mixpanel, we do a lot of work with segment, so a lot of CDP-related work, a lot of, you know, cohort testing segmentation.
30 00:04:09.700 ⇒ 00:04:21.370 Uttam Kumaran: And, like, sort of watching conversion rates through different funnels. This could be for marketing, this could be for product. Again, we work with some B2B software companies, we work with some e-com folks.
31 00:04:21.380 ⇒ 00:04:30.780 Uttam Kumaran: We work with some retail people, so most of their stuff is really focused on advertising conversions. But yeah, a lot of stuff on… on interaction data, for sure.
32 00:04:32.040 ⇒ 00:04:36.319 Elizabeth Gilbert: Cool. And, it sounds like you have a relatively small team. Are there…
33 00:04:36.440 ⇒ 00:04:47.030 Elizabeth Gilbert: Opportunities you’re looking for right now that do primarily that, or is it, some of that and some of, selecting tools and, configuring, kind of workspace and data?
34 00:04:47.240 ⇒ 00:04:53.310 Uttam Kumaran: Yeah, I mean, anyone we bring on, we sort of expect to be able to work in, like, multiple parts of the data
35 00:04:53.440 ⇒ 00:04:56.240 Uttam Kumaran: you know, ecosystem. I think…
36 00:04:56.350 ⇒ 00:05:08.539 Uttam Kumaran: we’ve found it kind of difficult to bring on people that are, like, I… all I do is data engineering, because we’re not a huge consultancy, and I also don’t think that many customers need, like.
37 00:05:08.610 ⇒ 00:05:25.989 Uttam Kumaran: the amount of people that you typically would need to run, like, some data functions. I think if you have really stellar people that can sort of bridge the gap between a couple different areas, like, that’s what we go for. So, the folks who are analysts on our team, they also do a little bit of modeling. They also can present to a client. We have data engineers who also
38 00:05:25.990 ⇒ 00:05:32.440 Uttam Kumaran: do pipeline work, but then do modeling work. So I would say my background as well is, like, kind of full stack.
39 00:05:32.650 ⇒ 00:05:43.020 Uttam Kumaran: data, so we’re sort of looking for folks that are doing that. I think as we grow, we will have clients where it could be just, like, tons of time just in one area.
40 00:05:43.260 ⇒ 00:05:56.340 Uttam Kumaran: But we’re… we’re, like, kind of coming into clients where there are active fires, and so we just… we have engagements that’s usually, like, anywhere from 2 to 3 people from our team that are, like, running into whatever the data fire is, so…
41 00:05:56.520 ⇒ 00:06:05.860 Uttam Kumaran: it is sort of, like, whatever the client needs. But of course, like, we understand that there is different functions and depth within each role in the data team.
42 00:06:06.440 ⇒ 00:06:08.289 Elizabeth Gilbert: Totally. Yeah, that makes a lot of sense.
43 00:06:08.910 ⇒ 00:06:17.910 Uttam Kumaran: Cool. So yeah, I guess I would be interested in, like, hearing about your background, if anything, like, lines up with, sort of, the work that we’re doing, and if it would be of any interest.
44 00:06:18.500 ⇒ 00:06:30.270 Elizabeth Gilbert: Totally. So, I am most recently a product data scientist. I have been working at Duo, which is part of Cisco, on their endpoint security products suite.
45 00:06:30.610 ⇒ 00:06:42.610 Elizabeth Gilbert: And working embedded in… with product engineering and design folks, to kind of tell the story of their products overall, 3 to 6 teams, depending on, what time. So…
46 00:06:42.610 ⇒ 00:06:49.570 Elizabeth Gilbert: Also have experience with Amplitude, that was the tool we brought on during my time at Duo. Dbt…
47 00:06:49.570 ⇒ 00:07:03.619 Elizabeth Gilbert: SQL Python, PEX was the BI tool we used, visualization, just kind of that full-stack storytelling and creating, self-serve resources that help empower that team, as well as the centralized resources, like, consolidated tables and dbt.
48 00:07:04.390 ⇒ 00:07:23.810 Elizabeth Gilbert: defining and communicating metrics and things like that. So, that’s my kind of bread and butter. I come from startups before that, so, was at a five-person pre-seed startup as a data provider company, and then a 1500-person Series H that was, healthcare administration automation. So a variety of, experience, and
49 00:07:23.810 ⇒ 00:07:32.200 Elizabeth Gilbert: Has kind of just, continued to lead in this, product data and user interaction direction, so a lot of experience now with,
50 00:07:32.200 ⇒ 00:07:51.899 Elizabeth Gilbert: kind of a funnel of, adoption at a, product level, or, what customers and users are using a product more and more, and I’m really curious to dig even more granular at a, like, feature level, right now. So, that’s kind of the direction I’m going in, and, where I’ve come from.
51 00:07:52.490 ⇒ 00:08:11.519 Uttam Kumaran: Yeah, so I guess tell me, like, what’s next, you think, in your career? Like, do you think you’d go back in industry? I mean, of course, like, I think the challenge and fun of a company like ours is, like, we have everybody working on, kind of, multiple clients, so we have… of course, and we have a variety of clients, although we do have B2B SaaS products, we also have
52 00:08:11.640 ⇒ 00:08:31.279 Uttam Kumaran: like, omni-channel retail, e-commerce, like, online pharmacies, all of which use data across more than just digital, like, this may be for inventory, ERP, finance, so I guess, like, kind of give me a sense of, like, are you really focused on just product interaction data? Would that be interesting? But also, like.
53 00:08:31.320 ⇒ 00:08:41.189 Uttam Kumaran: we are definitely different than going back into a classic B2B startup, or B2B SaaS startup, so, like, if that… if… maybe talk to me a little bit about if that’s interesting.
54 00:08:41.990 ⇒ 00:08:53.319 Elizabeth Gilbert: Yeah, I’m most interested right now in kind of following this really specific curiosity and getting to continue to, work with, yeah, that user interaction data and, build on the expertise that I’ve
55 00:08:53.320 ⇒ 00:09:04.439 Elizabeth Gilbert: built at Duo. I think from there, after I kind of follow that curiosity, I think that would… I’d be, well-equipped for something like this in the future. But…
56 00:09:04.570 ⇒ 00:09:15.479 Elizabeth Gilbert: kind of have this, you know, there’s places where you can’t do A-B testing, like security, we couldn’t be changing what the users are seeing on a day-to-day basis. Yeah. And I’m really curious to be in an industry and place where you can.
57 00:09:15.480 ⇒ 00:09:25.909 Elizabeth Gilbert: And having done industry transitions before, I am excited and open to getting to lean into the curiosity in the new industry. So, that’s kind of where I’m going right now. I really just want to focus on that.
58 00:09:25.910 ⇒ 00:09:28.340 Elizabeth Gilbert: Specifically, and then…
59 00:09:28.790 ⇒ 00:09:48.229 Elizabeth Gilbert: who knows? After that, I think, that would set me up well for… I’ve… I have with the, we collaborated really closely with the analytics engineers on our team, so as a data scientist, I have, dbt and kind of end-to-end experience that, also brings in data literacy and really, kind of empowering the one-to-many, relationship with, stakeholders.
60 00:09:48.230 ⇒ 00:09:52.659 Elizabeth Gilbert: and their confidence in using their own data and understanding it. So,
61 00:09:52.820 ⇒ 00:09:59.410 Elizabeth Gilbert: yeah, I think, I think it could be something to broaden in the future, but, continue following the curiosity specifically in the short term.
62 00:09:59.740 ⇒ 00:10:16.260 Uttam Kumaran: Cool, yeah, I mean, most of our clients, again, this is where I think we also differ a little bit from, like, classic industry, is that we’re reporting directly to the CEO, CMO, COO, and these are typically, like, again, 20, 50, 100 million dollar organizations. So, for us, we’re, like.
63 00:10:16.300 ⇒ 00:10:25.349 Uttam Kumaran: they’re asking us a problem, and we’re asking us a question, and we’re finding any means to turn around. So the urgency is definitely a lot higher than what I…
64 00:10:25.460 ⇒ 00:10:27.839 Uttam Kumaran: I mean, I worked in startups my whole career.
65 00:10:28.880 ⇒ 00:10:33.070 Uttam Kumaran: But I would say these are better problems, because in startups, they’re almost…
66 00:10:33.140 ⇒ 00:10:52.140 Uttam Kumaran: Sometimes there’s, like, limited user data, or, like, the interactions aren’t there. In fact, these are organizations where they have a ton of data, and they have all the tooling maybe set up, or we’re setting it up, but they’re really just hamstring by no one’s asked and answered really fundamental questions about how should we price, how should we do user segmentation, what are the core funnel drop-offs?
67 00:10:52.250 ⇒ 00:11:04.800 Uttam Kumaran: And so we have active clients where myself or my business partner, we’re… we’re the most, kind of, like, senior people on the analysis side of the fence at the company, so we’re having to do that. So we’re basically trying to…
68 00:11:04.870 ⇒ 00:11:15.640 Uttam Kumaran: Sort of bring on somebody that can ideally start to take some of that weight off of us, and can be, you know, client-facing, can present those insights, gather the next
69 00:11:15.700 ⇒ 00:11:25.919 Uttam Kumaran: sort of, question to dig at. Again, but all of our stuff is pretty classic, like, we do… we do a lot of amplitude mixed panel work, so it’s a lot of sessionized interaction event data.
70 00:11:26.820 ⇒ 00:11:27.540 Elizabeth Gilbert: Totally.
71 00:11:28.130 ⇒ 00:11:43.170 Uttam Kumaran: So I guess, like, tell me, like, if, like, is that something that you’d be interested in? Like, we have, like, typically the way we start, and… and if you’d like to move forward, I’m happy to… we would definitely introduce you to more folks on our data team. But we typically have people start on, like.
72 00:11:43.230 ⇒ 00:11:52.119 Uttam Kumaran: Like, one client, and sort of pick off a little bit of work, and see what it’s like to kind of work with each other, and then sort of grow from there.
73 00:11:52.810 ⇒ 00:12:00.669 Uttam Kumaran: we… we tend to hire a little bit differently than most companies in that I… I’ve done a lot of technical interviews in my career.
74 00:12:00.710 ⇒ 00:12:17.220 Uttam Kumaran: And I don’t think there’s anything that… that is better than just, like, actually on the job, like, learning what it’s like to work within a data team, work with our clients and our culture, and so that’s, like, what I buy stores. I don’t think we’re particularly great at, like, doing technical case scenarios, although we do have
75 00:12:17.220 ⇒ 00:12:27.780 Uttam Kumaran: that process as well. So that’s, like, typically our MO. I guess I would love to know from your side if that’s interesting, or if you have, like, a timeline you’re looking at for, like, a next opportunity, or… or what do you think?
76 00:12:28.700 ⇒ 00:12:48.209 Elizabeth Gilbert: Yeah, I think this could be good, to, like, stay in touch and, could be a really good fit down the road, after I’ve, gotten the, like, really specific, next curiosity, that I’m interested in primarily, and then, yeah, kind of just get to consider,
77 00:12:48.440 ⇒ 00:12:54.509 Elizabeth Gilbert: a more broad, possibility here. Okay, cool. But yeah, I’d love to stay in touch for that.
78 00:12:54.510 ⇒ 00:13:01.800 Uttam Kumaran: Okay, cool, so… so I guess, like, you’re not… you’re not, like, looking for currently a new role right now, like, at a consultancy like ours, or I guess…
79 00:13:02.960 ⇒ 00:13:07.389 Uttam Kumaran: I’m just trying to clarify, like, should you want to get intro to more people on the team, or without.
80 00:13:07.390 ⇒ 00:13:07.820 Elizabeth Gilbert: Absolutely.
81 00:13:07.820 ⇒ 00:13:09.179 Uttam Kumaran: Short term, or… yeah.
82 00:13:09.180 ⇒ 00:13:13.220 Elizabeth Gilbert: Currently not looking for a role that is as broad, more specific.
83 00:13:13.220 ⇒ 00:13:13.850 Uttam Kumaran: Okay.
84 00:13:13.850 ⇒ 00:13:31.300 Elizabeth Gilbert: So, that’s where I think it could be a good fit for the future. I am currently searching, but, with the layoff that I was at Cisco in August, they gave us plenty of severance, so I can search for the really specific thing first, and then potentially in a next phase, if I need to, broaden.
85 00:13:31.780 ⇒ 00:13:32.520 Elizabeth Gilbert: Yeah.
86 00:13:32.880 ⇒ 00:13:34.760 Elizabeth Gilbert: But that’s where I’m at right now, yeah.
87 00:13:34.760 ⇒ 00:13:48.020 Uttam Kumaran: Okay, yeah, well, let me know. I mean, again, we’re doing pretty… pretty awesome work for a bunch of clients in a bunch of different industries, so if that ends up being of interest, like, happy to consider you. And then, yeah, if I can be helpful at any time during your search or anything, please let me know.
88 00:13:48.550 ⇒ 00:13:50.590 Elizabeth Gilbert: Yeah, I really appreciate that. Thank you so much.
89 00:13:50.590 ⇒ 00:13:52.479 Uttam Kumaran: Okay, perfect. Well, appreciate the time today.
90 00:13:52.980 ⇒ 00:13:54.119 Elizabeth Gilbert: Yeah, great to meet you.
91 00:13:54.120 ⇒ 00:13:55.529 Uttam Kumaran: Okay, thank you. Bye.
92 00:13:55.760 ⇒ 00:13:56.360 Elizabeth Gilbert: Yeah.