Meeting Title: Brainforge x CTA: Kickoff! Date: 2025-11-14 Meeting participants: Katherine Bayless, Robert Tseng, Uttam Kumaran
WEBVTT
1 00:00:11.190 ⇒ 00:00:12.200 Katherine Bayless: Good morning.
2 00:00:12.720 ⇒ 00:00:13.600 Robert Tseng: Morning!
3 00:00:13.770 ⇒ 00:00:14.670 Katherine Bayless: How are you?
4 00:00:14.830 ⇒ 00:00:16.229 Robert Tseng: I’m good, how are you?
5 00:00:16.520 ⇒ 00:00:24.339 Katherine Bayless: Good. I’m gonna attempt to put in my headphones. They have not been playing nice, so if they don’t work, come back to regular.
6 00:00:24.340 ⇒ 00:00:26.150 Robert Tseng: Sure, sure, no worries, take your time.
7 00:00:26.810 ⇒ 00:00:27.570 Katherine Bayless: Yeah.
8 00:00:49.260 ⇒ 00:00:49.940 Katherine Bayless: Cheers.
9 00:00:50.420 ⇒ 00:00:51.649 Katherine Bayless: Stick to regular.
10 00:00:51.960 ⇒ 00:00:52.770 Robert Tseng: Okay.
11 00:00:52.770 ⇒ 00:00:58.849 Katherine Bayless: I think it’s a Lenovo, like, driver problem, because they’ve been working fine, and then just, like, a week ago, they…
12 00:00:59.630 ⇒ 00:01:00.050 Robert Tseng: Hmm.
13 00:01:00.050 ⇒ 00:01:03.529 Katherine Bayless: Been having other trouble with Zoom and these laptops that can’t connect.
14 00:01:03.530 ⇒ 00:01:04.220 Robert Tseng: Yeah.
15 00:01:05.050 ⇒ 00:01:06.440 Katherine Bayless: Hello, good morning.
16 00:01:06.440 ⇒ 00:01:07.560 Uttam Kumaran: Hey, good morning!
17 00:01:07.650 ⇒ 00:01:10.620 Katherine Bayless: How’s everything going? It’s good. How are you?
18 00:01:10.860 ⇒ 00:01:24.429 Uttam Kumaran: Good, good to see you. Glad to finally kick off. We’ve been thinking about all the stuff that we originally talked about, and I know the event is coming up even sooner than we talked about before, so…
19 00:01:24.430 ⇒ 00:01:43.849 Uttam Kumaran: Happy to, dive into things. Maybe as I’ll give a brief introduction. Robert is my business partner here at Brave Forge, so both of us, you know, run a good amount of the business together. Kind of similar background to me, except more on the analysis, side, and on the marketing side, so…
20 00:01:43.990 ⇒ 00:01:49.079 Uttam Kumaran: just wanna introduce you to him, but I’ll probably be the core interface for now, and then…
21 00:01:49.160 ⇒ 00:02:04.539 Uttam Kumaran: we have a pretty deep bench of folks, so wherever we need support, you know, as we discover things, you know, we can pull people in. So, yeah, just wanted to have Robert here just to kind of get a lay of the land today. Yeah.
22 00:02:04.540 ⇒ 00:02:13.470 Uttam Kumaran: And so, I have a… you know, I have a little bit of an agenda, but I’m happy to just follow your lead if you had a couple things you wanted to go through, or if you want to give a…
23 00:02:13.730 ⇒ 00:02:15.700 Uttam Kumaran: Kind of update on where things are at.
24 00:02:16.290 ⇒ 00:02:21.189 Katherine Bayless: Yeah, let me, let me try to sort out my thoughts here.
25 00:02:21.240 ⇒ 00:02:39.869 Katherine Bayless: So, I will say one thing, I will try to hop off probably around 8.45, another call at 9, I just need a little wiggle room to get in that brain space. So, in terms of updates, I’ll just do a couple high-level ones, and then we’ll work through your agenda, because it’s probably better than my random ramblings.
26 00:02:39.870 ⇒ 00:03:04.750 Katherine Bayless: So, SDG, the other consulting firm that we were also bringing in, they’re here. They got, their stuff through, like, I think it was… actually, it might have been two weeks ago now, so they’re here and in the systems and doing their thing. That’s actually who I have the call with at 9, just to kind of touch base and see how the first two weeks went. Their first, sort of, you know, marching orders are to do a really rapid, light-touch
27 00:03:04.750 ⇒ 00:03:22.690 Katherine Bayless: kind of data assessment of our Power BI deployment, mostly because I want to have some data to point to as I say we should move off Power BI, and here’s how. And so that’s what they’re going to be working on initially, and then we’ll kind of go through the next phases with them.
28 00:03:22.690 ⇒ 00:03:34.200 Katherine Bayless: you guys are the Snowflake folks, they’re very excited, for your work to get up and running. And then the other, team that we’re trying to bring on is AWS Professional Services, and,
29 00:03:34.550 ⇒ 00:03:46.859 Katherine Bayless: just as a side quest story, our legal team put the AWS Terms of Service in the packet for signatures, and so AWS said they couldn’t sign it. I was like, but they’re your terms?
30 00:03:49.740 ⇒ 00:03:53.010 Katherine Bayless: Like, I’m gonna lose my mind. I’m like, I know I get it, I’m not a lawyer, but anyway.
31 00:03:53.950 ⇒ 00:04:01.760 Katherine Bayless: So, we are waiting for… Okay. …to finish through the work, so you are number two to make it.
32 00:04:01.760 ⇒ 00:04:02.100 Uttam Kumaran: Right.
33 00:04:02.100 ⇒ 00:04:06.999 Katherine Bayless: process. But, yeah, so that’s all still in flight.
34 00:04:07.010 ⇒ 00:04:31.099 Katherine Bayless: yes, the event is starting to crush everything. Like, this is my first lap around the sun with this organization. I feel like everybody tried to warn me that by November, things have just been taken over, and they are very correct. And so, I am… honestly, my focus is mostly on desperately trying to get these contracts through, so that people can start working, and then just trying to stay out of the way.
35 00:04:31.100 ⇒ 00:04:31.820 Uttam Kumaran: Okay.
36 00:04:31.820 ⇒ 00:04:35.689 Katherine Bayless: So, one last update,
37 00:04:35.910 ⇒ 00:04:39.710 Katherine Bayless: One that is relevant to you guys and your work. So, Snowflake.
38 00:04:41.390 ⇒ 00:04:56.340 Katherine Bayless: So I, I launched an instance through the AWS Marketplace maybe about a month or so ago, just because I was like, well, why not at least start playing around? Definitely having a lot of fun in there. I think there are some things we could rapidly build out, like.
39 00:04:56.430 ⇒ 00:05:01.990 Katherine Bayless: probably by Tuesday, honestly. Yeah. There’s some low-hanging fruit for you guys.
40 00:05:02.830 ⇒ 00:05:05.829 Katherine Bayless: But we ran into a snag last week,
41 00:05:06.800 ⇒ 00:05:20.119 Katherine Bayless: I asked our VP of IT, who you’ll get to know, if he could bring it into the SSO and set up SCIM, because I think I want to have a lot of the business users able to access it, even if it’s just, like, some dashboards and streamlit stuff like that.
42 00:05:20.120 ⇒ 00:05:20.530 Uttam Kumaran: Right.
43 00:05:20.920 ⇒ 00:05:22.410 Katherine Bayless: He bricked it.
44 00:05:22.820 ⇒ 00:05:24.200 Uttam Kumaran: Oh, really? Okay, what happened?
45 00:05:26.020 ⇒ 00:05:30.179 Katherine Bayless: He let Claude do it, and the first thing Claude did was drop all users on the database.
46 00:05:30.180 ⇒ 00:05:32.969 Uttam Kumaran: Oh, yeah, great, okay.
47 00:05:32.970 ⇒ 00:05:46.480 Katherine Bayless: I think he reached out to support. Support said, well, you guys are pay-as-you-go, so tough bananas, and so I think we might just have to start over, which is honestly not the biggest deal. Lessons learned, and maybe you can help him?
48 00:05:46.480 ⇒ 00:05:51.639 Uttam Kumaran: I can… I can help, and I… I feel like I can get you guys,
49 00:05:52.180 ⇒ 00:05:55.089 Uttam Kumaran: Or, yeah, I’ll have a couple other ways.
50 00:05:55.090 ⇒ 00:06:13.770 Katherine Bayless: Yeah. I mean, was there anything in there? Did you end up loading anything in there, or… So, not… I mean, really, no, honestly. I had connected the S3 stage so that I could get into our data lake, and then I pulled in a couple little things just to make sure it worked, but I mean, in terms of delivering anything that’s in production, no.
51 00:06:13.770 ⇒ 00:06:14.650 Uttam Kumaran: Okay, okay.
52 00:06:14.870 ⇒ 00:06:19.540 Katherine Bayless: But really fucking cool platform. Like, not that I didn’t know that, but I even.
53 00:06:19.540 ⇒ 00:06:26.050 Uttam Kumaran: No, it’s changed a lot in the last, 12 to 15 months. Like, they completely made it, like, more of a…
54 00:06:26.180 ⇒ 00:06:33.399 Uttam Kumaran: workspace versus just, like, it used to just be, like, every BI tool, where it’s just, like, or warehouse tool, where it’s just a SQL editor.
55 00:06:33.400 ⇒ 00:06:33.890 Katherine Bayless: Yeah.
56 00:06:33.890 ⇒ 00:06:41.620 Uttam Kumaran: They made it so you can do a lot. So yeah, I guess that’s where I think you… you now have the luxury to, like, make a couple decisions on how to centralize things.
57 00:06:41.980 ⇒ 00:06:45.470 Uttam Kumaran: There’s a lot you can do just within the Snowflake platform, you know.
58 00:06:45.470 ⇒ 00:06:59.629 Katherine Bayless: Yeah, yeah, exactly. And I think, like, the friendliness towards a business user with the ability to put stuff in, like, a streamlit dashboard with that gorgeous, yeah, like, I mean, shut up and take my money. So…
59 00:06:59.630 ⇒ 00:07:04.800 Uttam Kumaran: A lot of our clients are not taking advantage of that, because they’re just used to, like, the old ways.
60 00:07:05.010 ⇒ 00:07:22.000 Uttam Kumaran: Of, like, doing BI and dashboards, and so one of the things that we’re starting to try to promote across the board is, like, trying to implement out of the box, like, chat with data use cases. You know, because a lot of your users, of course, will be doing lookups and basic things, but, like.
61 00:07:22.330 ⇒ 00:07:29.080 Uttam Kumaran: you’re starting this when that’s available, which is great, you know, and Snowflake, like, can power a lot of that out of the box.
62 00:07:29.300 ⇒ 00:07:37.409 Katherine Bayless: Yeah, no, I do have to, like… that is the silver lining to the chaos here, right? It’s like, you know, blank slate is blank slate.
63 00:07:37.410 ⇒ 00:07:38.039 Uttam Kumaran: Thank you guys.
64 00:07:38.040 ⇒ 00:07:43.970 Katherine Bayless: I don’t have anything to start with, but it also means I can make different choices from the ground up than I might have been able to otherwise, so…
65 00:07:43.970 ⇒ 00:07:44.450 Uttam Kumaran: Yeah.
66 00:07:44.450 ⇒ 00:08:01.340 Katherine Bayless: Yeah, never let a crisis go to waste. Yeah. Now, the other wrinkle, though, about Snowflake, and this is actually, it might just wind up being a very solvable problem. If not, it might be something where I want to bring in, like, just the firepower of knowledge from your side to the table.
67 00:08:01.440 ⇒ 00:08:18.689 Katherine Bayless: So the… the origin story for us getting Snowflake was one of our vendors, which is called ReMembers, or Impexium was the former name, so this is our association management system. I don’t know if you’ve interacted with many AMS platforms in your adventures, but they tend to be underwhelming.
68 00:08:18.800 ⇒ 00:08:32.800 Uttam Kumaran: Okay. And Nixium is actually one of the better ones. I’ve worked with them on two previous implementations. Like, their platform’s pretty good, but they’re definitely not in a build phase, they’re in a get more money out of existing clients kind of phase. Okay.
69 00:08:32.799 ⇒ 00:08:45.910 Katherine Bayless: So they had launched, instead of using our janky APIs for integrations, why don’t you purchase our data share via Snowflake, right? So this is kind of how we got to Snowflake so quickly. Otherwise, it would have been a later decision, I think.
70 00:08:46.720 ⇒ 00:08:52.970 Katherine Bayless: Well… I got, like, 3 weird emails from them earlier this week, because we finally got that contract through.
71 00:08:53.120 ⇒ 00:09:00.640 Katherine Bayless: like, asking for information piece by piece, and I was like, this doesn’t inspire confidence, but I wasn’t expecting to have confidence.
72 00:09:00.860 ⇒ 00:09:16.539 Katherine Bayless: Finally, it’s, can we get on a call? And now it’s their chief technology officer wants to talk to me on Monday, because they are hosting their Snowflake instance in Azure in, I guess, you know, the equivalent of US East 1, and they don’t have cross-cloud fulfillment enabled.
73 00:09:16.540 ⇒ 00:09:17.790 Uttam Kumaran: Oh, really? Why not?
74 00:09:18.740 ⇒ 00:09:20.880 Uttam Kumaran: Not a question for you, but yeah.
75 00:09:21.360 ⇒ 00:09:24.500 Uttam Kumaran: Yeah. Alright.
76 00:09:24.500 ⇒ 00:09:35.289 Katherine Bayless: Those are literally the words that fell out of my mouth, too, guys. I was like, what? Like, what? I don’t understand. Like, what? So I think… and I sent an email around to the membership team yesterday.
77 00:09:35.290 ⇒ 00:09:38.079 Uttam Kumaran: What is the name of that company, by the way?
78 00:09:38.080 ⇒ 00:09:47.469 Katherine Bayless: Re, colon members, like, remembers, but with a semi-fole in the middle. They used to be called Impexium, I-M-P-E-X-I-U-M.
79 00:09:47.650 ⇒ 00:09:48.210 Uttam Kumaran: Okay.
80 00:09:49.650 ⇒ 00:10:01.579 Katherine Bayless: It really, it’s not a big deal, like, that’s the crazy part to me, but I know these companies, and I know these… specifically these people, and they’re totally gonna try to be like, well, just put your snowflake in Azure.
81 00:10:01.580 ⇒ 00:10:05.890 Uttam Kumaran: And I’m like, nope. Yeah, so you should… we can help… we can help figure that out.
82 00:10:05.890 ⇒ 00:10:09.620 Katherine Bayless: Yeah, I mean, I’m not overly worried about it, but I do think…
83 00:10:09.620 ⇒ 00:10:12.290 Uttam Kumaran: I would also, like, I assume, like, yeah, I guess…
84 00:10:13.210 ⇒ 00:10:18.870 Uttam Kumaran: There’s other ways for them to do that, to get that data to us, like, several other ways.
85 00:10:18.870 ⇒ 00:10:19.310 Katherine Bayless: Yeah.
86 00:10:19.310 ⇒ 00:10:23.549 Uttam Kumaran: And, yeah, they should also just enable cross-platform.
87 00:10:24.760 ⇒ 00:10:33.780 Katherine Bayless: I mean, we’re literally paying them $10,000 a year just to have access to the data share, and I’m like.
88 00:10:33.970 ⇒ 00:10:45.199 Uttam Kumaran: Yeah, but I just don’t know why they… I mean, that’s what I want to ask them, and I mean, I’ll even… I could send them to docs, because we’ve… I’ve done this before. We tried to host some data
89 00:10:45.630 ⇒ 00:10:48.469 Uttam Kumaran: For some of the clients, we were procuring and
90 00:10:48.590 ⇒ 00:10:58.989 Uttam Kumaran: curating some data, so I’m like, oh, we should just put it up in the data marketplace. So I went through the process of hosting and sharing data, and you could totally do this. It’s a setting, yeah.
91 00:10:58.990 ⇒ 00:11:08.359 Katherine Bayless: Yeah, yeah. No, and I asked Eric, the sales guy from Snowflake, who’s been really lovely to talk to, and he said the same thing. He’s like, I mean, they can do it, so…
92 00:11:08.360 ⇒ 00:11:09.480 Uttam Kumaran: Yeah, okay.
93 00:11:09.480 ⇒ 00:11:24.089 Katherine Bayless: I think, honestly, AMS companies tend to prey upon the lack of technical knowledge at their clients, right? Because it’s a lot easier for them to get on that call on Monday and say, oh, too bad, just host your Snowflake in Azure and it’ll work, and for me to go, oh, okay.
94 00:11:24.090 ⇒ 00:11:25.660 Uttam Kumaran: Okay, yeah, yeah, yeah, yeah, yeah.
95 00:11:25.810 ⇒ 00:11:38.910 Uttam Kumaran: But I don’t… I don’t fuck around like that. Yeah, good, good. So, yeah, okay, so yeah, I mean, I would say for any of these vendors, you have us to just toss in, or completely be, like, own…
96 00:11:39.120 ⇒ 00:11:47.099 Uttam Kumaran: own this getting done. Similarly for Snowflake, like, who is the… who is the person you’re interfacing? I just have a lot of…
97 00:11:47.740 ⇒ 00:11:51.170 Uttam Kumaran: past and current contacts at Snowflake, so I’m happy to just, like.
98 00:11:51.690 ⇒ 00:11:52.980 Katherine Bayless: Yes, and…
99 00:11:52.980 ⇒ 00:12:17.969 Katherine Bayless: The guy, it’s, Matt Crush, young guy, seems really… he’s really interesting. Like, he wanted to go into politics, but then was like, this is boring, and so he’s like, I’ll just go to tech sales, and I’m like, okay, I really need to understand that way better. But, so he’s been reaching out, because he’s been obviously trying to get us to go the contract, like, pre-purchase credits route, right? And they have non-profit discounts, which I was actually surprised are available for us each
100 00:12:17.970 ⇒ 00:12:20.980 Katherine Bayless: Normally, only C3s get the discounts.
101 00:12:20.980 ⇒ 00:12:33.030 Katherine Bayless: But he said that they are… we would be eligible. I did float it past our finance team. I think they’ve seen me set $300,000 on fire in, like, a month, and they’re like, girl.
102 00:12:33.370 ⇒ 00:12:38.960 Uttam Kumaran: Yeah, but also, you don’t need to do… you don’t need to buy… these guys, this is what they’ll do. They’ll just keep snowflake books.
103 00:12:38.960 ⇒ 00:12:39.530 Katherine Bayless: I know.
104 00:12:39.530 ⇒ 00:12:45.200 Uttam Kumaran: Yeah, so it’s brutal, but, like, just go month to month for a while. Yeah.
105 00:12:45.200 ⇒ 00:13:04.970 Katherine Bayless: That’s kind of what I’ve told him, is like, I… eventually, I think a contract will make sense, right? Like, why leave money on the table once we have a better sense of our spend? But I was like, I really need 6 months of data to just understand how correct I am in my hypothesis around adoption before I start making bets, because I don’t want to bet too small, and I don’t want to bet too big. So yeah, he’s really lovely to work with.
106 00:13:05.590 ⇒ 00:13:08.029 Katherine Bayless: But yeah, that’s who I’ve been talking to.
107 00:13:08.030 ⇒ 00:13:08.700 Uttam Kumaran: Okay, great.
108 00:13:09.270 ⇒ 00:13:16.779 Katherine Bayless: So yeah, there’s, like, 10,000 other, you know, updates and things, but those are the most salient pieces, so we can dive into the agenda if you’re ready.
109 00:13:17.120 ⇒ 00:13:28.530 Uttam Kumaran: Okay, cool. I mean, yeah, we also… I guess, overview of, like, kind of, like, what we wanted to try to get done is, one is, like, Snowflake dbt, I think. Second was…
110 00:13:28.670 ⇒ 00:13:29.900 Uttam Kumaran: Maybe I’ll go and…
111 00:13:30.380 ⇒ 00:13:36.249 Uttam Kumaran: I’ll kind of go in this order, it’s like, second is, like, Power BI, sort of, wherever we can help on…
112 00:13:36.580 ⇒ 00:13:48.810 Uttam Kumaran: either migrating, or making a decision, or, like, kind of just, like, observing that. And then the third piece is, like, the company and people identity resolution piece.
113 00:13:48.990 ⇒ 00:14:05.820 Uttam Kumaran: I would say maybe we can… I kind of get the world around Snowflake, and maybe we can talk a little bit about where you need us on the Power BI stuff, if… if anything, and then we can talk… I think the rest of the call will mainly talk through the…
114 00:14:06.150 ⇒ 00:14:09.240 Uttam Kumaran: Company, person, and then we talk access and things.
115 00:14:09.240 ⇒ 00:14:17.890 Katherine Bayless: Okay, cool. So… Okay, literally, you just said those words, and my brain went, you know.
116 00:14:18.910 ⇒ 00:14:38.270 Katherine Bayless: Oh, the Power BI, thank you, yes, yes, yes, yes, great. Right, coffee. Okay, so Power BI. So, SDG, I think, so they’re very much like an Accenture type, right? Like, they’re just big, spongy consult… I mean, like, on the kickoff call, they’re like, can we get your org chart? And I’m like, no, you cannot make sales calls to my coworkers yet. Please calm down.
117 00:14:39.610 ⇒ 00:14:44.100 Katherine Bayless: So, but they’re really good, and I… so I want them to…
118 00:14:44.120 ⇒ 00:15:00.300 Katherine Bayless: I want them to go through and just kind of, like, what’s in there, honestly? Like, I think, organically, at this point, I’ve been able to gather enough information about, like, what are the things people slack me with angry face about? You know, what are the places where I really am a blocker to information?
119 00:15:00.300 ⇒ 00:15:08.819 Katherine Bayless: And so, I think while they’re kind of doing the assessment piece, not really any need for you guys to get involved, very cut and dry work. What I would like to kind of…
120 00:15:08.890 ⇒ 00:15:19.720 Katherine Bayless: be sharing learnings and collaborating more is they’re going to, at the end of it, give us, like, basically a couple prototypes in other BI platforms, right?
121 00:15:19.790 ⇒ 00:15:44.339 Katherine Bayless: in theory, they could recommend staying on Power BI. I doubt that will be the recommendation. But… so they were going to mock some stuff up. I think they were going to do Sigma, which was one Snowflake had recommended, Omni, which you guys had recommended. But I would really like that to be something that, like, the full group is looking at, right? Because you guys are going to have a totally different angle on what our BI platform should be going forward.
122 00:15:44.340 ⇒ 00:15:50.349 Katherine Bayless: based on the Snowflake and entity resolution work, then they will, looking at the graveyard of questions we’ve asked in the past.
123 00:15:50.350 ⇒ 00:16:04.009 Uttam Kumaran: And so, I really think there is power in that brain trust of coming at what BI platform makes sense for us, or BI approach, right? Like, I don’t even know if it’s going to be a platform anymore, potentially. No, I agree, and that’s where, like, I would say we…
124 00:16:04.330 ⇒ 00:16:10.769 Uttam Kumaran: for a lot of our clients, we just want to share that we know all of the players, but I… I actually…
125 00:16:10.810 ⇒ 00:16:35.099 Uttam Kumaran: only want to make the decision that, like, actually drives adoption. So you don’t want to have self-serve analytics, as nobody is self-serving, and it’s a couple of analysts producing. The other situation is, like, you don’t want to have a Power BI or Tableau if you have a lot of developers, because the developer experience really sucks. So it’s going to be purely based on, like… I mean, one is, for us, like, we just go kind of meet all the…
126 00:16:35.270 ⇒ 00:16:41.529 Uttam Kumaran: all the characters, and that is what really… on the BI side, for me, that’s what drives the decision.
127 00:16:41.650 ⇒ 00:16:57.339 Uttam Kumaran: You know, also, like, again, for some of our clients, they’re, like, they want to then… they want to use this and then eventually embed it for their customers, so there’s… yeah, so there’s, like, a great decision to make there. And then, of course, it’s pricing. And so that’s where, across the board.
128 00:16:57.510 ⇒ 00:17:10.049 Uttam Kumaran: We just are helpful in those vendor negotiations, because it’s just a small world in data, and usually, again, you’re right, it’s that they just prey on, like, they find out whatever you don’t know, and kind of, like, go deep.
129 00:17:10.119 ⇒ 00:17:23.069 Uttam Kumaran: Versus, like, I just… this is all we do is buy data software, so… kind of, like, don’t… you kind of can’t, like, really mess with us on many stuff, and we just go and we drive the best discounts, the best terms.
130 00:17:23.260 ⇒ 00:17:26.010 Uttam Kumaran: And then we kind of… so that’s where I think…
131 00:17:27.020 ⇒ 00:17:35.000 Uttam Kumaran: that is a good opportunity to throw us into the fire, where it’s, like, very annoying to deal with, like, more vendors and things like that, so…
132 00:17:35.140 ⇒ 00:17:42.210 Katherine Bayless: Yeah, I… that’s like a sort of, you know, asterisk sidequest note. I think next year…
133 00:17:42.450 ⇒ 00:17:46.899 Katherine Bayless: It’s interesting, the conversations I’ve been trying to seed don’t seem to go as…
134 00:17:47.150 ⇒ 00:18:12.110 Katherine Bayless: I would expect them to, but there is a lot of room for vendor consolidation and renegotiation at this organization. We have 3 CVEN platforms, right? Like, that’s insane, and I have to have an access to all three, because I’m the data person, and right, you know, so there’s just… there’s a lot of room for improvement. I think people don’t understand what I mean when I say that yet, though. Okay. That’s like a 2026 education project around, like, this is how much we are spending, because when.
135 00:18:12.110 ⇒ 00:18:20.390 Uttam Kumaran: And that’s exactly it, right? Like, we did this for a couple vendors, where it’s like, oh, you have, like, MixedPanel and Amplitude, or you have…
136 00:18:20.390 ⇒ 00:18:32.510 Uttam Kumaran: a couple of CRMs, and of course, like, the vendor doesn’t care, you bought it in isolation, and there… yeah, 100%, at least you can negotiate better terms, and then second is…
137 00:18:32.660 ⇒ 00:18:40.270 Uttam Kumaran: you could drive towards just one platform, but yeah, okay, that makes sense. Yeah, I mean, that’s just, like, sitting… sitting money, right?
138 00:18:40.270 ⇒ 00:18:40.660 Katherine Bayless: Right.
139 00:18:40.660 ⇒ 00:18:41.620 Uttam Kumaran: Right.
140 00:18:41.620 ⇒ 00:18:44.700 Katherine Bayless: Right, right. So, yeah, I think,
141 00:18:44.700 ⇒ 00:19:08.630 Katherine Bayless: Yeah, next year, lots of… lots of opportunities. But yeah, so I think really with the interplay between the two groups, it’s like a friendly red teaming of recommendations, right? Like, you guys are gonna have the color outside the lines perspective, they’re gonna have the, we’re bored because we’ve done this assessment 10 million times, and here’s the recommendation we give everybody. Somewhere in between there is the right answer for us, and so that’s kind of how I envision that collaboration working.
142 00:19:08.680 ⇒ 00:19:15.119 Katherine Bayless: And also, if there’s other tools you want to recommend them to build prototypes on, I’m all ears. Omni and Sigma look really fucking cool, though.
143 00:19:15.410 ⇒ 00:19:33.640 Uttam Kumaran: Yeah, Omni and Sigma are great. I would, like, the… I would say Sigma is now, like, becoming to, like, the super, super enterprise level. Omni is still, like, I think, innovating a lot, especially with their chat with, like, data features. But there’s also a lot of other kind of tools that are coming up. I mean.
144 00:19:33.730 ⇒ 00:19:39.259 Uttam Kumaran: Again, another tool we try to figure out early on is just something for observability and, like, alerting.
145 00:19:39.750 ⇒ 00:19:55.739 Uttam Kumaran: So, like, we… we’ve used Metaplane in the past, but there’s a couple of folks that we can… we can look at. So, yeah, I mean, but then otherwise, it’s, like, usual suspects around something for version control, something for Snowflake, so Snowflake, and then something for… for dbt.
146 00:19:56.290 ⇒ 00:20:04.320 Uttam Kumaran: Okay, cool, so then… so that makes sense, kind of on Snowflake, and then, yeah, in Snowflake, basically, we just kind of do the work, so it’s…
147 00:20:04.960 ⇒ 00:20:23.059 Uttam Kumaran: RBAC, like, setting up warehouses, service accounts, like, I guess, is there anything sort of outside the… and I can give you, like, what our typical plan is, and we will actually save all those grants and stuff on your end, so you have them, because that’s kind of, like, the worst to backtrack.
148 00:20:23.510 ⇒ 00:20:24.030 Katherine Bayless: Yeah.
149 00:20:24.030 ⇒ 00:20:25.629 Uttam Kumaran: Is there anything else that’s, like.
150 00:20:25.900 ⇒ 00:20:28.370 Uttam Kumaran: Particular in there that we need to…
151 00:20:29.510 ⇒ 00:20:45.890 Katherine Bayless: So, that would be awesome, actually. I was gonna ask, like, if you have kind of, like, a high-level playbook where I can go through and be like, okay, this is kind of what we’re gonna… Yeah. Yeah, yeah. So specifically, this was, like, one of those middle-of-the-night, like, oh, type moments. With the RBAC,
152 00:20:46.110 ⇒ 00:20:53.440 Katherine Bayless: it occurs to me that the actual, like, setting up and administering of it, I trust you guys to teach me how.
153 00:20:53.610 ⇒ 00:21:01.780 Katherine Bayless: I’ll be honest, I have not really thought through what that should look like. This organization right now has really strict.
154 00:21:01.780 ⇒ 00:21:10.629 Uttam Kumaran: our back, but it’s organic, right? It’s just whether or not Susie will give you access to her spreadsheet on SharePoint, right? So, like, I need to kind of figure out…
155 00:21:10.630 ⇒ 00:21:16.839 Katherine Bayless: What are going to be the heuristics around access for data at this company in a way
156 00:21:17.020 ⇒ 00:21:21.920 Katherine Bayless: that makes sense, is scalable, and also that I can sell.
157 00:21:21.920 ⇒ 00:21:40.160 Katherine Bayless: Because I’m kind of team… generally, my, you know, I have the naive view of information, right? I think more is better, and so, like, my initial thought was just, like, yeah, data to the people. And I think the low-hanging fruit stuff that we can build will be fine for anybody to be able to see at the organization, because it’s the kind of stuff that’s already general.
158 00:21:40.160 ⇒ 00:21:47.330 Katherine Bayless: But I do think there might be more nuance and sensitivity around some of the data than I think I was really…
159 00:21:47.680 ⇒ 00:21:53.540 Uttam Kumaran: Who is the… who is, like, the core person, like, in that world? Or, like…
160 00:21:53.870 ⇒ 00:21:58.779 Uttam Kumaran: It does… who needs, like, Who’s driving, sort of, security, or…
161 00:21:59.380 ⇒ 00:22:03.739 Uttam Kumaran: vulnerability into, like, audit logs or things like that. Is there somebody internally?
162 00:22:07.750 ⇒ 00:22:09.630 Katherine Bayless: So,
163 00:22:11.720 ⇒ 00:22:21.579 Katherine Bayless: On paper, the answer is yes. We have a virtual CISO through a company called Hartman Advisors, which are a consulting shop in the association space, so that tells you everything you need to know about them.
164 00:22:21.600 ⇒ 00:22:22.280 Uttam Kumaran: Okay.
165 00:22:22.280 ⇒ 00:22:28.430 Katherine Bayless: So we have a part-time guy who’s basically retired, who occasionally says things about security.
166 00:22:28.720 ⇒ 00:22:29.170 Uttam Kumaran: Okay.
167 00:22:29.170 ⇒ 00:22:45.050 Katherine Bayless: internally, we have my friend Jay, our VP of IT, who bricked our Snake instance this week. He is actively trying to almost, kinda, sorta, re-implement Okta, but basically, like, the do-it-live version, so.
168 00:22:45.050 ⇒ 00:22:45.400 Uttam Kumaran: Okay.
169 00:22:45.400 ⇒ 00:22:49.669 Katherine Bayless: He reset everybody’s passwords, and it did not go well.
170 00:22:50.750 ⇒ 00:23:02.889 Katherine Bayless: But we really do not have security posture at this organization. Like, I mean, I shouldn’t say that, it is a recorded call and all the things, but, like, we are… it is… it’s bad. It’s really, really bad. And so there’s a part of me that’s, like.
171 00:23:02.890 ⇒ 00:23:13.659 Katherine Bayless: okay, I need to not add to the cowboy problems, right? Like, I also don’t want to come in with the draconian security approach and make everybody just go, like.
172 00:23:13.660 ⇒ 00:23:22.600 Uttam Kumaran: No, so yeah, I mean, on our world, it’s exactly… so when we set up Snowflake, it’s all, like, role-based, and so you can then assign roles
173 00:23:22.800 ⇒ 00:23:29.679 Uttam Kumaran: to users, or roles or groups, and so across all of the tools we’ll use, we will take that. And so.
174 00:23:29.810 ⇒ 00:23:32.879 Uttam Kumaran: At the… it’ll be more, like, functional roles.
175 00:23:32.880 ⇒ 00:23:34.150 Katherine Bayless: Like…
176 00:23:34.150 ⇒ 00:23:44.019 Uttam Kumaran: Do you have access to read or write on what? And then we can then move in. In the BI layer, I feel like, and again, it depends on how many people are going to be accessing Snowflake.
177 00:23:44.020 ⇒ 00:23:54.489 Uttam Kumaran: And the BI layer is where you’ll have groups, and then you’ll have roles, and then you’ll have, like, content folders. And we can map… we can map all of that out, and so… and then…
178 00:23:54.490 ⇒ 00:23:59.659 Katherine Bayless: Not only is that one piece for implementing, of course you need a process for…
179 00:23:59.660 ⇒ 00:24:14.219 Uttam Kumaran: requesting access that’s audited, or at least goes some… is sitting somewhere, and then a grant is run. So we can do, like… I mean, it’s actually great, I think, that you’re thinking about it. We can just do the lightweight of, like.
180 00:24:14.240 ⇒ 00:24:25.759 Uttam Kumaran: okay, everything’s set up in a very flexible, role-based way. Here is a clear area for you to look at who’s accessing what, and here’s a way for people to request additional. So you want to have…
181 00:24:25.760 ⇒ 00:24:40.680 Uttam Kumaran: So, typically, with very little friction, you want to have, like, least access principle, right? And then… and then… but you don’t… but I think… I think that’s actually a good place to start, because then you can just… you can actually grant the accesses, as long as they’re just logged somewhere, and then…
182 00:24:40.760 ⇒ 00:24:46.379 Uttam Kumaran: Eventually, maybe someone comes in and thinks about security across the board, and your team is, like, in the best…
183 00:24:46.460 ⇒ 00:25:06.300 Uttam Kumaran: in the best shape, maybe, or at least you can do some type of monthly audit of who has access to what. Yeah. You know, the way… the way this… and Snowflake is really the biggest place to get this right. It’s just hard to roll back roles, so you just don’t want to grant, like, all privileges to all users and sort of call it a day.
184 00:25:06.560 ⇒ 00:25:11.010 Katherine Bayless: I should say, in terms of, like, making it not sound too terrifying.
185 00:25:11.010 ⇒ 00:25:11.460 Uttam Kumaran: Fair.
186 00:25:11.460 ⇒ 00:25:26.130 Katherine Bayless: Like, so the… in terms of access in an administer, sort of context, right, like, the people that will be working, working in Snowflake, that’s a small group that we can well define. It is the business layer piece that I’m like, oh god.
187 00:25:26.130 ⇒ 00:25:50.869 Katherine Bayless: Okay. And I think some of it, too, is because our staff… I mean, hi, hello, the call we’re on right now, a lot of our staff are not actually our staff, right? Like, so we are a very vendor-dependent organization. This is one of the things Jay struggles with that I do have empathy for. It’s like, the number of people that are acting on our network at any given time, I mean, it’s, like, three times the size of our staff, and he doesn’t usually find out until somebody says, like, hey.
188 00:25:50.870 ⇒ 00:25:53.490 Katherine Bayless: Let somebody, you know, or let so-and-so access this.
189 00:25:53.490 ⇒ 00:25:56.179 Uttam Kumaran: Anyone needs something, yeah, what are they doing here? Yeah, yeah, yeah.
190 00:25:56.180 ⇒ 00:26:15.900 Katherine Bayless: Right, so I think it’s governance on that business layer that I’m just like… because I think my data-to-the-people attitude will run into some friction with, like, oh, but I’ve never been able to see that data before, or, you know, they’ve never been able to see our data, kind of thing. There’s just a lot of analog silos running around.
191 00:26:16.440 ⇒ 00:26:20.870 Uttam Kumaran: So there’s something on the security piece. Okay, and then, yeah, maybe let’s,
192 00:26:21.810 ⇒ 00:26:32.459 Uttam Kumaran: let’s talk a little bit about, the, like, company and people stitching, right? So that’s, I feel like, kind of one of the big holy grail.
193 00:26:33.130 ⇒ 00:26:33.909 Katherine Bayless: Yes, dude.
194 00:26:33.910 ⇒ 00:26:38.970 Uttam Kumaran: And maybe we can… I do have some notes on, sort of, what we talked about, which is, like.
195 00:26:39.360 ⇒ 00:26:45.640 Uttam Kumaran: We talked about ingesting from various sources, sort of, like, source of truth for companies.
196 00:26:45.920 ⇒ 00:26:55.729 Uttam Kumaran: I’m assuming there’s… yeah, we… I can talk at length about, sort of, doing identity stitching, but maybe you can just start there, and I can take notes.
197 00:26:56.390 ⇒ 00:27:03.849 Katherine Bayless: Yeah, I think, it’s… it’s an interesting, like, spot. I…
198 00:27:04.150 ⇒ 00:27:07.880 Katherine Bayless: I’m not sure to what extent we might be able to, like.
199 00:27:08.060 ⇒ 00:27:19.100 Katherine Bayless: clean everything up real good and be good for a while, honestly, versus do we really need, like, a CDP or a tool that is purpose-built to, like, keep your data clean?
200 00:27:19.740 ⇒ 00:27:30.909 Katherine Bayless: I kind of think we’ll skew somewhere in the middle, but towards the, like, just clean it up and use your tools better angle. But we are… we’re definitely dying the death by a thousand cuts on this right now, so…
201 00:27:30.910 ⇒ 00:27:55.779 Katherine Bayless: for people, because we have no central system of record, I’m starting to sort of introduce the idea of, like, a canonical ID, master data management ID, right? I call it DataOps ID, and so I’m starting to kind of try to propagate that where I can, and sort of have an ID that’s following people around. Even though I know right now they’re not much better than email addresses, they eventually could be used for merge, right? You know.
202 00:27:56.180 ⇒ 00:27:58.260 Katherine Bayless: Like, at least I’m putting something out there.
203 00:27:58.400 ⇒ 00:27:59.750 Katherine Bayless: So…
204 00:28:00.190 ⇒ 00:28:21.019 Katherine Bayless: admittedly, and this could just be my, like, looking for familiar patterns, brain, like, I keep gravitating towards, I really need a table of, like, all of these data ops IDs I’ve put out there, and every other system ID I can figure out that it’s associated with, because we’re just, we’re running into constant, like, oh, well, this person’s over there, but not over here, and…
205 00:28:21.020 ⇒ 00:28:21.630 Uttam Kumaran: guy.
206 00:28:21.630 ⇒ 00:28:37.350 Katherine Bayless: their registration type needs to get changed in this system, but they’re a speaker over there, but they want to go by this name on their speaker listing, and this on their badge, and I’m like, one of my bad guys, like, how have you been doing this? Like, this is just chaos, right? And it’s all Excel spreadsheets and intern tiers.
207 00:28:37.440 ⇒ 00:28:42.210 Katherine Bayless: So, the people are a challenge, the companies…
208 00:28:42.650 ⇒ 00:28:52.129 Katherine Bayless: are an interesting challenge, right? Because a person can pick up the phone and call us and get angry that they can’t get into a system. Companies were kind of, like.
209 00:28:52.540 ⇒ 00:28:55.260 Katherine Bayless: Knowingly delivering a terrible experience to?
210 00:28:55.260 ⇒ 00:28:55.760 Uttam Kumaran: Yeah, yeah, yeah.
211 00:28:55.760 ⇒ 00:29:01.309 Katherine Bayless: flying blind about it. So, for example, the one thing that’s causing struggles
212 00:29:01.820 ⇒ 00:29:09.309 Katherine Bayless: when you go to register for CES, which we should circle back on that, did you know it takes 29 minutes and 36 seconds on average for someone to come?
213 00:29:09.310 ⇒ 00:29:13.679 Uttam Kumaran: I would like to go register and see to kind of see the process, yeah, so that’s…
214 00:29:13.680 ⇒ 00:29:30.100 Katherine Bayless: So we… if you’re… if you work for one of our member companies, you get to go for free, kind of a thing, right? As in a regular attendee. And so what we have to do is upload a list of the company’s domains to our registration site, so that when.
215 00:29:30.100 ⇒ 00:29:31.560 Uttam Kumaran: Hi, Liz? Yeah.
216 00:29:31.560 ⇒ 00:29:39.969 Katherine Bayless: Exactly. But these are… I mean, it’s Samsung, and Panasonic, and LG. They don’t have atlg.com and that’s it, right?
217 00:29:39.970 ⇒ 00:29:57.179 Katherine Bayless: g.co.jp and .co.uk, and, you know.mail, all of these things, right? And so, like, we get these onesie-twosies where the companies are like, well, we can’t, you know, our people can’t register, and it’s like, well, it’s because we didn’t, like, you know, allow list that one weird flavor of your domain.
218 00:29:57.180 ⇒ 00:29:57.830 Uttam Kumaran: Yeah, yeah, yeah.
219 00:29:57.830 ⇒ 00:30:08.569 Katherine Bayless: the domain research in the first place is manual, and sometimes the domain the company sends to CES is not really their main domain, it’s their research arm, or their innovation lab, or whatever, all these things, right?
220 00:30:08.570 ⇒ 00:30:09.270 Uttam Kumaran: Perfection.
221 00:30:09.270 ⇒ 00:30:20.730 Katherine Bayless: what happens is, membership will basically finally get somebody to buy membership, selling them on, but you can go to CES, and then they will try to go to CES, and registration won’t work, and then they get very mad at us, and then I…
222 00:30:20.730 ⇒ 00:30:23.029 Uttam Kumaran: So everything is, like, yeah, piecemeal.
223 00:30:23.030 ⇒ 00:30:23.910 Katherine Bayless: Right, right.
224 00:30:23.910 ⇒ 00:30:27.750 Uttam Kumaran: Yeah, I mean, so, I think certainly, I’m sure there’s…
225 00:30:27.920 ⇒ 00:30:30.489 Uttam Kumaran: Kind of, like, old-world things we can do.
226 00:30:30.490 ⇒ 00:30:33.630 Katherine Bayless: Right, yeah, exactly. I mean, like, old techniques here, like, yeah, we don’t need.
227 00:30:33.760 ⇒ 00:30:41.979 Uttam Kumaran: I would say also, like, look, we… within Snowflake, you have Cortex, you can start to do, like, sort of more, less deterministic matching.
228 00:30:41.980 ⇒ 00:30:44.290 Katherine Bayless: I think it’s something that…
229 00:30:44.410 ⇒ 00:30:51.630 Uttam Kumaran: I wish I had this type of technology when I was doing a lot of this type of identity stitching, building, like, canonical
230 00:30:51.680 ⇒ 00:31:02.139 Uttam Kumaran: org accounts, people, like, customers, data sets, but we should have… I mean, basically, this is just, like, kind of like waterfall joining, but I think
231 00:31:02.140 ⇒ 00:31:18.669 Uttam Kumaran: one thing that could be fun here is for us to try to say, like, hey, if this doesn’t fit any of our more deterministic criteria, maybe we should ask AI to run to see if it can do a match. Like, that technology, again, like, I don’t think many people are taking advantage of, but Snowflake allows you to run
232 00:31:18.820 ⇒ 00:31:23.820 Uttam Kumaran: gen AI directly in SQL.
233 00:31:23.820 ⇒ 00:31:24.390 Katherine Bayless: Yeah.
234 00:31:24.390 ⇒ 00:31:25.650 Uttam Kumaran: All of them, so we should…
235 00:31:25.650 ⇒ 00:31:26.120 Katherine Bayless: Yeah.
236 00:31:26.120 ⇒ 00:31:35.559 Uttam Kumaran: We should try then. So then… so, I mean, for us, like, if we were even to break down that example, kind of the way we work is, one, like, we just sort of build, like.
237 00:31:35.870 ⇒ 00:31:43.569 Uttam Kumaran: we start to build just, like, a big diagram of this whole thing. Like, we start to go through, we just build a diagram of all the core sources.
238 00:31:43.570 ⇒ 00:31:48.139 Katherine Bayless: we sort of get a stamp on, like, okay, we have everything here. We also identify, like.
239 00:31:48.140 ⇒ 00:31:53.300 Uttam Kumaran: who is, like, the owner of that source? Like, so that’s the person to go befriend.
240 00:31:53.550 ⇒ 00:32:11.369 Uttam Kumaran: And then it’s like, okay, we can talk a little bit about ETL, but even if we don’t arrive at, like, the automated solution, at least we can do manual exports, load stuff in, and then we can start modeling based on what we know the schema is, and then we just drive towards, like.
241 00:32:11.600 ⇒ 00:32:13.739 Uttam Kumaran: the, some type of marts.
242 00:32:14.190 ⇒ 00:32:14.620 Katherine Bayless: Yeah.
243 00:32:14.620 ⇒ 00:32:24.300 Uttam Kumaran: Where you have a source of truth for company and person, and then whatever, like, intermediate models need to support that.
244 00:32:24.650 ⇒ 00:32:39.720 Katherine Bayless: Yeah, I think… I think yes. Yeah, I mean, like, check, done, go, right? I think that is exactly where we need to start. I think where we might run into some interesting questions that become sort of more business questions than technical ones, really, is, like.
245 00:32:40.750 ⇒ 00:32:49.189 Katherine Bayless: Membership goes to the company, the, like, it has to be, I think, the USHQ for the company, and then it trickles down to, like, subsidiaries.
246 00:32:49.900 ⇒ 00:32:54.969 Katherine Bayless: That part sounds so clear, but then, like, the reality is it gets really messy.
247 00:32:54.970 ⇒ 00:33:02.470 Uttam Kumaran: No, no, no, this, like, account org structure is not very clear at all. Like, we have… we dealt with this at, I mean, every company we work with, where there’s, like.
248 00:33:02.760 ⇒ 00:33:08.259 Uttam Kumaran: Parent accounts, and then accounts, and then there’s, like, orgs, and account… yeah, so…
249 00:33:08.260 ⇒ 00:33:28.250 Katherine Bayless: Well, I think we aren’t super clear in our language around, like, what entities do we record in the database, right? You know what I mean? So, like, if a giant multinational company joins, do we ask them for all 1,000 subsidiaries? I mean, we probably should, but right now we don’t. We just kind of wait for the ones that matter to get mentioned.
250 00:33:28.250 ⇒ 00:33:28.600 Uttam Kumaran: Yeah.
251 00:33:28.600 ⇒ 00:33:41.540 Katherine Bayless: But, like, I think that’s a question the business is going to have to answer to a certain extent, is, like, what data do we want to be pulling in by default anytime, you know, we’re getting a new company? And then are we monitoring them for mergers and acquisitions and stuff like that?
252 00:33:41.540 ⇒ 00:33:58.269 Uttam Kumaran: Yeah, that’s something we can drive, like, we can show what the state of the world is, and then kind of do a little basic analysis of, like, here’s all these accounts, here’s sort of some of the relationships, and then kind of drive towards a decision on, like, yeah, this basically, like, parent org structure, you know.
253 00:33:58.270 ⇒ 00:33:58.690 Katherine Bayless: Yeah.
254 00:33:58.690 ⇒ 00:34:01.790 Uttam Kumaran: convention, and, like, how we organize this, so…
255 00:34:02.110 ⇒ 00:34:24.450 Katherine Bayless: Yeah, and I think, the other piece, too, that maybe just, like, in a recommendation kind of context, if, as part of the work, you would be able to maybe say, like, hey, based on the types of things that you’re looking at around companies, you know, we would recommend purchasing Dun & Bradstreet’s data, or… Right. Right, because, like, I think we’re going to want to purchase somebody’s data, I just don’t know whose.
256 00:34:25.030 ⇒ 00:34:37.259 Uttam Kumaran: Yeah, we also do a lot of support on enrichment, so we actually have another client where we’re basically doing the same dev, like, almost, like, they are offering enrichment to their customers, so we’re… we’ve evaluated almost, like.
257 00:34:37.370 ⇒ 00:34:39.840 Uttam Kumaran: We’re evaluating almost 10 or 15 different vendors.
258 00:34:39.840 ⇒ 00:34:40.750 Katherine Bayless: Okay, nice.
259 00:34:40.750 ⇒ 00:34:46.040 Uttam Kumaran: I think we’ve… we usually have… we’ve worked with the usual suspects, like DMV, Clearbit.
260 00:34:46.389 ⇒ 00:34:56.179 Uttam Kumaran: Apollo, so… yeah. And I think the other thing is some of these new vendors have really interesting data, so that’s what I would be interested to hear from sales.
261 00:34:56.359 ⇒ 00:35:00.379 Uttam Kumaran: Like, what are… what are, like,
262 00:35:00.569 ⇒ 00:35:08.459 Uttam Kumaran: wave a magic wand, what kind of data would I want on, like, leads? Because now there’s some vendors that are tracking job movements.
263 00:35:08.910 ⇒ 00:35:16.809 Uttam Kumaran: that are, of course, tracking capital raises, M&A, and so that’s really… I think that’s a great win for those
264 00:35:16.960 ⇒ 00:35:32.949 Uttam Kumaran: sales folks, which is like, hey, what new things? And, like, how do we… that… I think that’s kind of how, when we talk about meeting at all the characters, for me, it’s more about, like, what is a small win I can get for them to show that, like, we actually care, and then that will unlock
265 00:35:33.090 ⇒ 00:35:42.450 Uttam Kumaran: whatever, like, news I’m coming with the next meeting, like, I need something, or we need to set something up, or, like, we need… I need to talk to your… whoever’s
266 00:35:42.720 ⇒ 00:35:49.519 Uttam Kumaran: your salesperson for your vendor, like, so I’m trying to see, like, what are quick wins we can get for all those folks, so…
267 00:35:49.520 ⇒ 00:35:50.160 Katherine Bayless: Nice, okay.
268 00:35:50.280 ⇒ 00:35:56.130 Uttam Kumaran: Yeah, everyone is a piece, then totally, when I talk to them, I can ask them about… Babe.
269 00:35:56.480 ⇒ 00:36:03.270 Katherine Bayless: Yeah, okay, okay. Yeah, because I think… I think that would be cool, and lovely to benefit serendipitously from work you’re already doing.
270 00:36:03.270 ⇒ 00:36:12.279 Uttam Kumaran: Yeah, yeah, which, I agree. We’re just talking to, like, all these different vendors about all the data. We’re kind of profiling, so it’s really convenient.
271 00:36:12.280 ⇒ 00:36:14.259 Katherine Bayless: Yeah, nice, nice, nice.
272 00:36:14.260 ⇒ 00:36:15.290 Uttam Kumaran: Okay.
273 00:36:15.290 ⇒ 00:36:26.450 Katherine Bayless: I think it really… I do think old-school techniques are gonna yield a lot of improvement in our data, so I think starting there. And I guess, question in terms of sequencing.
274 00:36:27.150 ⇒ 00:36:41.220 Katherine Bayless: knowing that there might be some delays on getting the Impexium data, I mean, we’ll see what the call goes like on Monday, but also knowing that we haven’t quite gotten the AWS crew in yet to set up the new accounts.
275 00:36:42.710 ⇒ 00:36:45.080 Katherine Bayless: And Jay broke my existing Snowflake instance.
276 00:36:45.170 ⇒ 00:37:04.749 Katherine Bayless: do we want to spin up another Snowflake in the current environment, just to give you guys a place to start working? Or do we want to, like, I’ll invite you into our old Postgres database so we can do the entity resolution type stuff for a bit, and see if, like, the AWS stuff moves rapidly enough to just start Snowflake in the new account?
277 00:37:04.750 ⇒ 00:37:09.659 Katherine Bayless: I guess really the question is, how much benefit would it be to not have to move Snowflake?
278 00:37:10.540 ⇒ 00:37:12.349 Uttam Kumaran: Yeah,
279 00:37:15.660 ⇒ 00:37:21.170 Uttam Kumaran: It is not, like, not very easy to move Snowflake across… Yeah. Okay.
280 00:37:21.660 ⇒ 00:37:33.450 Uttam Kumaran: So, the alternative… I mean, basically, if we’re, like, still, like, okay, let’s start in Postgres, we can just locally model stuff with, like, DuckDB, and basically try to just
281 00:37:33.860 ⇒ 00:37:35.520 Uttam Kumaran: Produce some outputs.
282 00:37:35.790 ⇒ 00:37:46.830 Uttam Kumaran: I mean, if you… if you, like, yeah, I guess the… the biggest, it’s a tough question.
283 00:37:47.410 ⇒ 00:37:49.549 Katherine Bayless: It’s fine if you want to take the weekend to think about it, too.
284 00:37:49.550 ⇒ 00:37:59.019 Uttam Kumaran: Yeah, I think also, like, maybe if you want, if, like, if you could invite me to the existing one, does it just… oh, you can’t even log in or anything. And Matt didn’t help with that?
285 00:37:59.500 ⇒ 00:38:20.539 Katherine Bayless: He, well, I mean, to be fair to him, he did say, theoretically, support is for people that have a contract, right? You know, like, fair. But he did say that if the account admin role wasn’t deleted, you should be able to get back in, and I was like, I don’t think we deleted roles, I think Jay just deleted all the users, and so it’s like, all the… he said all the SCIM users are in, but the SAML isn’t working.
286 00:38:20.540 ⇒ 00:38:22.740 Uttam Kumaran: Can you send… do you know your account ID?
287 00:38:22.740 ⇒ 00:38:23.350 Katherine Bayless: Yeah.
288 00:38:23.670 ⇒ 00:38:26.000 Uttam Kumaran: Okay, send me that, and then I will…
289 00:38:26.000 ⇒ 00:38:29.919 Katherine Bayless: I do, I do, I do. Do I know it without getting in? Yes, I do. Okay, hang on.
290 00:38:29.920 ⇒ 00:38:36.679 Uttam Kumaran: If you have your account ID, and then your region, or, like, any other identifying information, I can,
291 00:38:36.850 ⇒ 00:38:38.980 Uttam Kumaran: Let me just, like, ask some people.
292 00:38:39.340 ⇒ 00:38:45.400 Uttam Kumaran: Because also, if we’re coming in as, like, a partner, like, and we’ve done a lot of snowflake work.
293 00:38:45.540 ⇒ 00:38:47.530 Uttam Kumaran: I’m sure they can help us with something.
294 00:38:47.870 ⇒ 00:38:51.000 Katherine Bayless: Yeah, well, and I think, too, like, I’m genuinely…
295 00:38:51.310 ⇒ 00:38:56.940 Katherine Bayless: This particular instance, if it has to just, like, toss that in the trash and start over, really, like.
296 00:38:56.940 ⇒ 00:38:57.440 Uttam Kumaran: Okay, okay.
297 00:38:57.440 ⇒ 00:39:00.999 Katherine Bayless: anything that was like, I’ll never be able to rebuild.
298 00:39:01.200 ⇒ 00:39:02.120 Uttam Kumaran: Okay, okay.
299 00:39:02.120 ⇒ 00:39:05.460 Katherine Bayless: Why am I not finding where I put this?
300 00:39:11.490 ⇒ 00:39:16.260 Katherine Bayless: Oh yeah, here we go, again. Okay, so there’s an account…
301 00:39:17.500 ⇒ 00:39:22.210 Katherine Bayless: locator, as well as an account ID, is that right?
302 00:39:22.370 ⇒ 00:39:24.219 Uttam Kumaran: Yeah, you should have both.
303 00:39:24.220 ⇒ 00:39:24.980 Katherine Bayless: Okay.
304 00:39:26.080 ⇒ 00:39:27.420 Katherine Bayless: Yep, okay.
305 00:39:30.110 ⇒ 00:39:33.770 Katherine Bayless: I’ll just drop them right here in the Zoom chat for the ease of the moment.
306 00:39:34.910 ⇒ 00:39:38.279 Katherine Bayless: And then the locator is…
307 00:39:41.940 ⇒ 00:39:47.869 Katherine Bayless: The region, I’m assuming, is US East 2. It doesn’t look like I wrote that down in my little…
308 00:39:48.080 ⇒ 00:39:48.650 Uttam Kumaran: Okay.
309 00:39:48.650 ⇒ 00:39:52.300 Katherine Bayless: grabbing the info, but I launched it from US East 2 in AWS, if that…
310 00:39:52.300 ⇒ 00:39:52.900 Uttam Kumaran: Okay.
311 00:39:53.820 ⇒ 00:39:56.249 Katherine Bayless: Me, necessarily, they’d be the same.
312 00:39:56.510 ⇒ 00:39:59.850 Katherine Bayless: But yeah, it’s really… it’s not a problem to start a fresh account.
313 00:39:59.850 ⇒ 00:40:00.390 Uttam Kumaran: Yeah, okay.
314 00:40:00.390 ⇒ 00:40:10.330 Katherine Bayless: like, really, it’s fine. And then, like, you know, ideally, my, you know, my hope had been that we would have these sequenced so that AWS would be in, and they’d open the account, and then…
315 00:40:10.330 ⇒ 00:40:10.850 Uttam Kumaran: Yeah, good.
316 00:40:10.850 ⇒ 00:40:12.440 Katherine Bayless: start, but yeah.
317 00:40:12.440 ⇒ 00:40:15.210 Uttam Kumaran: No, I mean, we’re gonna keep driving, no matter what.
318 00:40:15.210 ⇒ 00:40:15.760 Katherine Bayless: Wait a minute.
319 00:40:15.760 ⇒ 00:40:19.139 Uttam Kumaran: Gonna keep driving, and so, yeah.
320 00:40:19.270 ⇒ 00:40:20.010 Katherine Bayless: Yeah.
321 00:40:20.350 ⇒ 00:40:39.349 Uttam Kumaran: So, I know we have 5 minutes, let me, let me just run through a couple, like, maybe logistics, and yeah, I have a thousand more questions, but I will keep sending them to you, and we can hop back on. So, I guess, like, one is we try to just, like, touch base at least once a week, like, I would love to just, like, put time on, and then we…
322 00:40:39.350 ⇒ 00:40:42.279 Uttam Kumaran: Internally, we’ll talk about y’all every day.
323 00:40:42.280 ⇒ 00:40:58.329 Uttam Kumaran: happy to… if you want to join that, or whatever, no matter… whatever we want to get to, I’m sure you’re very busy. And so, I think, one, is there a particular day that, like, works best if we were to put, like, something recurring on? For us, it’s just to make sure that we
324 00:40:58.440 ⇒ 00:41:03.180 Uttam Kumaran: Like, say hi once a week, not through virtual means, you know?
325 00:41:03.180 ⇒ 00:41:28.120 Katherine Bayless: Right, yeah, no, I think, so I am… I am delinquent in setting up a stand-up, for the other team. We had talked about it, and picked day 9. I think we’d split the difference at 9.15, to do a stand-up, so if that works for you guys as well, I can invite everybody to that daily stand-up, and I can finally do my homework and actually send it. But in terms of, like, is there a day… I mean, weirdly, Monday
326 00:41:28.120 ⇒ 00:41:32.739 Katherine Bayless: kind of tend to be better, because everybody’s panicking about everything else before they.
327 00:41:32.740 ⇒ 00:41:33.240 Uttam Kumaran: Alright.
328 00:41:33.240 ⇒ 00:41:34.110 Katherine Bayless: I mean?
329 00:41:34.330 ⇒ 00:41:38.330 Katherine Bayless: Fridays had been pretty open, but now they’re getting a little tight.
330 00:41:38.330 ⇒ 00:41:38.910 Uttam Kumaran: Okay.
331 00:41:38.910 ⇒ 00:41:49.260 Katherine Bayless: But I really… I… as much as I know my time is compressed and quickly stolen, like, this is the most important thing for me to be working on, so I wanted…
332 00:41:49.260 ⇒ 00:42:00.709 Uttam Kumaran: We’re not going to be hoarding too much of it, it’s just helpful. Some stuff is hard to do async, and like, I just want to make sure we have the block, and then we’ll… we’re open to talk whenever we need to talk.
333 00:42:00.710 ⇒ 00:42:02.109 Katherine Bayless: Yeah, yeah.
334 00:42:02.110 ⇒ 00:42:04.959 Uttam Kumaran: Okay, so, maybe, like, I mean.
335 00:42:05.270 ⇒ 00:42:10.690 Uttam Kumaran: Yeah, I would like to do… usually we do, like, Thursdays, because every… most of our companies are, like.
336 00:42:10.810 ⇒ 00:42:16.840 Uttam Kumaran: they’re kind of rough with Mondays, but up to you if you prefer beginning of the week or end of week.
337 00:42:17.350 ⇒ 00:42:27.659 Katherine Bayless: Yeah, I mean, Thursday’s fine, if that is what you guys want to do. Thursdays, I have, like, an 11 and a 2 that are, like, recurring calls. Okay. But…
338 00:42:27.790 ⇒ 00:42:34.940 Katherine Bayless: I mean, honestly, also, the 98 and 9am spots are good, because most folks won’t do meetings before, like, 10 here.
339 00:42:34.940 ⇒ 00:42:35.480 Uttam Kumaran: Okay.
340 00:42:35.480 ⇒ 00:42:38.879 Katherine Bayless: So, if you don’t mind hitting me early, then that also helps.
341 00:42:38.880 ⇒ 00:42:42.490 Uttam Kumaran: Okay, so let’s… maybe let’s plan for Thursday mornings, and we’ll kind of see how it goes.
342 00:42:42.490 ⇒ 00:42:44.129 Katherine Bayless: Okay, okay, yeah, yeah.
343 00:42:44.130 ⇒ 00:42:44.730 Uttam Kumaran: Cool.
344 00:42:45.320 ⇒ 00:42:53.749 Uttam Kumaran: That way it gives us, like, a… what I like to do Thursday, so it gives us a day in case, like, there’s something quickly that we can move forward, we have that, versus, like.
345 00:42:53.910 ⇒ 00:42:56.209 Uttam Kumaran: Oh, it’s a weekend now, so…
346 00:42:56.210 ⇒ 00:43:06.669 Katherine Bayless: Yeah. Well, and I think maybe given the, you know, so many things, going on and to talk about, like, if you guys have some time on Monday and you want to do, like, another hour versus.
347 00:43:06.670 ⇒ 00:43:07.110 Uttam Kumaran: Yeah.
348 00:43:07.110 ⇒ 00:43:16.560 Katherine Bayless: of just deep dive, like, that, I think, would make a lot of sense to me. And I might ask if the PM, at least, from the SDG folks, would like to join for a portion, maybe? Yeah.
349 00:43:17.200 ⇒ 00:43:28.730 Katherine Bayless: Maybe I’ll schedule a separate call for that, actually, now that I’m thinking about that. Let’s you and I meet Monday. Let me see if I’m getting a big group thing later in the week. Okay. Yeah. Yeah.
350 00:43:33.400 ⇒ 00:43:35.190 Katherine Bayless: Your dog’s so cute.
351 00:43:35.700 ⇒ 00:43:38.470 Uttam Kumaran: He woke up early, too.
352 00:43:38.630 ⇒ 00:43:39.560 Katherine Bayless: Hmm.
353 00:43:39.890 ⇒ 00:43:45.690 Uttam Kumaran: Okay, and then I guess my next question was gonna be about, like, access and comms.
354 00:43:45.850 ⇒ 00:43:49.969 Uttam Kumaran: Whatever, whichever, pick your voice in.
355 00:43:49.970 ⇒ 00:43:57.629 Katherine Bayless: Yeah, so I think, communications-wise, I mean, Zoom, Slack, email, email is mostly…
356 00:43:57.630 ⇒ 00:44:00.860 Uttam Kumaran: Slack works, okay, so we can, we can totally do a Slack workspace, I can invite you.
357 00:44:00.860 ⇒ 00:44:01.390 Katherine Bayless: land.
358 00:44:01.390 ⇒ 00:44:01.910 Uttam Kumaran: Okay.
359 00:44:01.910 ⇒ 00:44:12.399 Katherine Bayless: Okay, and then we use Asana for project management. I am similarly delinquent in really building anything out in there, to be honest. We desperately need a ticketing system, really.
360 00:44:12.400 ⇒ 00:44:27.580 Uttam Kumaran: And we use linear, we’re happy to do it in Asana, or… yeah, whatever you think. Like, we will take out all of our stuff out in our linear instance, no matter what. So we can give you access to that, or if you want it to all kind of get consolidated.
361 00:44:27.580 ⇒ 00:44:29.010 Katherine Bayless: In Asana, we can…
362 00:44:29.680 ⇒ 00:44:31.320 Uttam Kumaran: Kind of clone it there, too.
363 00:44:31.640 ⇒ 00:44:34.029 Katherine Bayless: Yeah… I need to think on that one.
364 00:44:34.030 ⇒ 00:44:34.840 Uttam Kumaran: Okay.
365 00:44:34.840 ⇒ 00:44:35.640 Katherine Bayless: Yeah.
366 00:44:35.790 ⇒ 00:44:39.109 Katherine Bayless: I will get back to you. And then…
367 00:44:39.110 ⇒ 00:45:00.990 Katherine Bayless: Oh, and then access. Yeah, so I have… because I know my focus is in 10,000 places, Kyle from the team is kind of just… I asked him if he wouldn’t mind being the grunt work of the logistics piece, so I’ll connect you with him. Okay. He can take care of the access-y type stuff. I’m thinking… I mean, all he’s gonna do is connect you to Jay, to be honest. Okay. But, like, in terms of what you guys want first…
368 00:45:00.990 ⇒ 00:45:07.810 Katherine Bayless: Probably getting you into the AWS account and GitHub, would make the most sense, the starting places. Okay, okay.
369 00:45:07.810 ⇒ 00:45:11.940 Uttam Kumaran: And then, yeah, if you already have a repo set up for a data platform, or if.
370 00:45:11.940 ⇒ 00:45:12.580 Katherine Bayless: Okay.
371 00:45:12.580 ⇒ 00:45:16.089 Uttam Kumaran: Or I can… I can tell Jay that that’s exactly what we need to start
372 00:45:16.290 ⇒ 00:45:22.669 Uttam Kumaran: putting… writing some stuff down, and then the AWS instance, I assume, is, like, all the Postgres data that you mentioned.
373 00:45:22.670 ⇒ 00:45:23.150 Katherine Bayless: Hmm.
374 00:45:23.150 ⇒ 00:45:23.740 Uttam Kumaran: N.
375 00:45:23.900 ⇒ 00:45:41.159 Katherine Bayless: You’ll actually… so, in AWS, you’ll see we have a Postgres data warehouse, which is the one that I’ve started and, you know, been sort of using for sanity. I also now have migrated the SQL Server from the old marketing team. Database migration service is really fun and easy to use, like.
376 00:45:41.160 ⇒ 00:45:42.430 Uttam Kumaran: Oh, really? Okay, great.
377 00:45:42.430 ⇒ 00:45:46.519 Katherine Bayless: I was like, okay, I ran into a really stupid little snag, but once I fixed that, I was like, this is a.
378 00:45:46.520 ⇒ 00:45:50.590 Uttam Kumaran: Was the old SQL server that wasn’t on AWS, or that was just going somewhere else? Okay.
379 00:45:50.590 ⇒ 00:46:09.710 Katherine Bayless: So it was Azure hosted, and there were, like, 6 or 7 databases on one server instance, I guess, and so I migrated all of them into one SQL server in the US. My next step had been to turn it into a Postgres instance, I just have not gotten there yet. So yes, you’ll see all of the data in there.
380 00:46:09.710 ⇒ 00:46:10.999 Uttam Kumaran: Okay, great, perfect.
381 00:46:11.000 ⇒ 00:46:18.699 Katherine Bayless: I’ll also point you to the S3 bucket, I’m pretty sure I just called it Data Lake, where I dumped all of the data out of the old SQL server into.
382 00:46:18.700 ⇒ 00:46:19.080 Uttam Kumaran: Right.
383 00:46:19.080 ⇒ 00:46:22.389 Katherine Bayless: Ian Parquette files for Snowflake, so that was what I had connected the old Snowflake.
384 00:46:22.390 ⇒ 00:46:24.419 Uttam Kumaran: Okay, okay, okay, perfect, okay, yeah.
385 00:46:24.750 ⇒ 00:46:35.369 Katherine Bayless: So yeah, and then in terms of GitHub, I do have a repo that I’ve been using to push infrastructure as code and whatnot back and forth with AWS. I’m not opposed to starting a fresh one for all.
386 00:46:35.370 ⇒ 00:46:50.950 Uttam Kumaran: I don’t know, we can just have… if you just want to create a data platform, it doesn’t matter. I would… I would prefer to have it… I mean, basically what we’re talking about is, like, dbt code, any, like, ad hoc analysis, docs, we try to push everything there, so… Okay. I don’t mind if we just use that, and…
387 00:46:50.950 ⇒ 00:46:51.360 Katherine Bayless: Okay.
388 00:46:51.360 ⇒ 00:46:53.170 Uttam Kumaran: Easy to split later. No, not a problem.
389 00:46:53.170 ⇒ 00:47:07.579 Katherine Bayless: Okay. Okay, okay, okay. Then, yeah, I will… I’ll email Kyle with you and Jay, and I’ll say we want AWS and GitHub, and then we’ll go from there on the rest. It’s not that you can’t have SharePoint and Power BI, I just feel like that is not the…
390 00:47:07.580 ⇒ 00:47:22.900 Uttam Kumaran: No, this is… this is enough, this is enough homework for us to poke at, and then, of course, we’ll talk on Monday, and so by… by sort of our meeting, I mean, maybe we can even arrive at, like, that. Like, how’s your Monday, like, afternoon, or how’s your Monday afternoons looking?
391 00:47:23.040 ⇒ 00:47:31.930 Katherine Bayless: Yeah, so by 3 o’clock, I will be free, so if you want to do a 3, or a 3.30 or something, or 4, whatever works, but yeah.
392 00:47:32.450 ⇒ 00:47:34.550 Uttam Kumaran: Okay, perfect. So I’ll just put a hold there.
393 00:47:38.810 ⇒ 00:47:47.989 Uttam Kumaran: Okay, yeah, that’ll give us enough time to just poke around, and then ideally on Monday, we could probably screen share and, like, walk through some of the data for a profile that we did.
394 00:47:48.500 ⇒ 00:47:52.359 Katherine Bayless: Yeah. Can I give you a tour of the, you know, method to the madness up here?
395 00:47:52.360 ⇒ 00:47:57.129 Uttam Kumaran: Yes. Well, I just kind of, like, want to start crossing, and we’re gonna build this diagram as we kind of go, so…
396 00:47:57.130 ⇒ 00:48:01.719 Katherine Bayless: Okay, great. One more thing I was gonna just, like, throw into the back of brain.
397 00:48:01.720 ⇒ 00:48:02.380 Uttam Kumaran: Sure.
398 00:48:02.670 ⇒ 00:48:19.999 Katherine Bayless: automated testing for data work and stuff like that, right? I mean, I… I’ve made a few funny boo-boos in my adventures so far that I’m like, okay guys, I have to be clear, I don’t normally make these mistakes, I just don’t normally work so alone, right? Yeah, yeah, yeah. And so, like, if there’s anything… we used Playwright at my last place.
399 00:48:20.000 ⇒ 00:48:20.470 Uttam Kumaran: Okay.
400 00:48:20.470 ⇒ 00:48:25.500 Katherine Bayless: the software engineering team, but I’m curious if there’s applications of that kind of stuff, and I think dbt does a little thing.
401 00:48:25.500 ⇒ 00:48:37.540 Uttam Kumaran: Yeah, dbt does a little bit, I mean, so kind of, we’ll totally have three different environments for development. We’ll have a development environment, we’ll have staging, so PRs will get tested in staging, then we’ll have production. So, there’s kind of…
402 00:48:37.540 ⇒ 00:48:51.590 Uttam Kumaran: multi-layers where, again, this is for us to decide as a team, but usually we’re like, run all your stuff in development, then push a PR. The PR has to kind of go through CI, and then those will be, like, automatic through dbt to check
403 00:48:51.590 ⇒ 00:49:03.070 Uttam Kumaran: not only run tests, but also run, like, compile all the models, execute, and then it’ll move to production. So that will… that will nix a bunch of issues making it to prod, and then…
404 00:49:03.150 ⇒ 00:49:13.310 Uttam Kumaran: Otherwise, on the actual production models, we can layer on Metaplane or another tool that will do monitoring. So let’s say we just want to make sure that, hey, at any moment, like.
405 00:49:13.370 ⇒ 00:49:28.330 Uttam Kumaran: the revenue column doesn’t go to zero, or, like, data isn’t stale. We can do those things in dbt. It’s just not, like, purpose-built for that. There’s a couple of tools that are kind of, like, we can trial and think about what we like. But at minimum.
406 00:49:28.490 ⇒ 00:49:35.359 Uttam Kumaran: we always do this, like, dev staging prod approach, because it just, like, catches so many issues at the PR level, so…
407 00:49:35.910 ⇒ 00:49:36.500 Katherine Bayless: Yeah.
408 00:49:36.500 ⇒ 00:49:41.919 Uttam Kumaran: Yeah, yeah. So we’ll put that… that’s, like, all part of, like, our dbt setup, basically.
409 00:49:41.920 ⇒ 00:49:42.880 Katherine Bayless: Okay, nice, nice.
410 00:49:42.880 ⇒ 00:49:58.590 Uttam Kumaran: So, when we set that up, you’ll be able to see that, and then… yeah, again, like, I want to… it should be like, hey, I pushed this PR, go into the repo, go into the warehouse to go check, like, the results, and then verify that the CI ran, and then we can merge it. So, yeah, and we can map that out as well.
411 00:49:58.870 ⇒ 00:50:00.829 Katherine Bayless: Cool. So excited.
412 00:50:00.830 ⇒ 00:50:05.130 Uttam Kumaran: Me too, this is great. We haven’t worked on, like, a super fresh…
413 00:50:05.680 ⇒ 00:50:09.980 Uttam Kumaran: usually we come in, it’s like… I mean, I can tell there is a lot of…
414 00:50:10.130 ⇒ 00:50:14.060 Uttam Kumaran: on fire, but I can also tell you, like, let’s just go, like.
415 00:50:14.060 ⇒ 00:50:14.560 Katherine Bayless: Yeah.
416 00:50:14.560 ⇒ 00:50:17.060 Uttam Kumaran: Come on, one by one, so that’s really helpful.
417 00:50:17.060 ⇒ 00:50:22.609 Katherine Bayless: Yeah, yeah. Before I go, really quickly, do you guys want to see where I’ve been falling from?
418 00:50:22.610 ⇒ 00:50:23.480 Uttam Kumaran: Yes.
419 00:50:25.080 ⇒ 00:50:26.460 Uttam Kumaran: No way!
420 00:50:26.460 ⇒ 00:50:28.930 Katherine Bayless: Okay, so the… I don’t know how well you can see it, really.
421 00:50:28.930 ⇒ 00:50:32.229 Uttam Kumaran: I was just like, yeah, maybe you’re in a… it’s like a hospital.
422 00:50:32.230 ⇒ 00:50:35.059 Katherine Bayless: Yeah, right. No, the pink one is mine.
423 00:50:35.650 ⇒ 00:50:37.020 Uttam Kumaran: Whaaat?
424 00:50:37.020 ⇒ 00:50:39.470 Katherine Bayless: This is weird.
425 00:50:39.470 ⇒ 00:50:40.519 Uttam Kumaran: Where is this?
426 00:50:40.520 ⇒ 00:50:45.799 Katherine Bayless: This Porsche Silver Spring, this is her break-in oil change, because I finally got through the break-in model.
427 00:50:45.800 ⇒ 00:50:47.489 Uttam Kumaran: What Porsche do you have?
428 00:50:47.490 ⇒ 00:50:49.289 Katherine Bayless: It’s a GTS 4-liter.
429 00:50:49.290 ⇒ 00:50:49.930 Uttam Kumaran: No…
430 00:50:49.930 ⇒ 00:50:50.500 Robert Tseng: move!
431 00:50:50.500 ⇒ 00:50:51.380 Uttam Kumaran: Ayy.
432 00:50:51.380 ⇒ 00:50:53.119 Katherine Bayless: It’s the dumbest thing I’ve ever done.
433 00:50:53.120 ⇒ 00:50:54.029 Uttam Kumaran: What the hell?
434 00:50:54.030 ⇒ 00:50:54.550 Katherine Bayless: Why not?
435 00:50:54.550 ⇒ 00:50:55.310 Uttam Kumaran: I know.
436 00:50:55.450 ⇒ 00:50:56.800 Katherine Bayless: Yeah,
437 00:50:57.600 ⇒ 00:50:58.230 Katherine Bayless: So yeah, I love it.
438 00:50:58.230 ⇒ 00:50:59.210 Uttam Kumaran: Awesome.
439 00:50:59.210 ⇒ 00:51:00.050 Katherine Bayless: Yeah.
440 00:51:00.050 ⇒ 00:51:04.400 Uttam Kumaran: When… have you always been into Porsches, or is that the, like… okay.
441 00:51:04.400 ⇒ 00:51:15.069 Katherine Bayless: I grew up around drag racers, but they had a very much, like, a cars- aren’t-for-girls kind of attitude, and I was, like, 34 years old before I realized that’s not actually a rule, and so I’m a late bloomer, but I’m here.
442 00:51:15.690 ⇒ 00:51:26.169 Uttam Kumaran: I’m so jealous. Yeah, I grew up so into cars, but have not just had the luxury to… I mean, I would love to buy, like, an old boxer or something. Yeah.
443 00:51:26.400 ⇒ 00:51:26.810 Katherine Bayless: Yes.
444 00:51:26.810 ⇒ 00:51:27.999 Uttam Kumaran: rent a car.
445 00:51:28.000 ⇒ 00:51:32.719 Katherine Bayless: I want, like, an old Corvette, like, from the, like, 70s, with those big wheel humps on it.
446 00:51:32.720 ⇒ 00:51:41.039 Uttam Kumaran: Yeah, yeah. Yes. Oh, amazing. Oh, that’s, like, that made my day, that’s awesome. You’ll have to send a picture of it, like, yeah, that’d be great.
447 00:51:41.040 ⇒ 00:51:42.750 Katherine Bayless: Totally, yes.
448 00:51:42.750 ⇒ 00:51:44.020 Uttam Kumaran: So yeah.
449 00:51:44.020 ⇒ 00:51:46.969 Katherine Bayless: I’m, like, half watching her out of the corner of my eye, like…
450 00:51:46.970 ⇒ 00:51:56.470 Uttam Kumaran: That’s great, that’s an awesome… that’s just an awesome place to work, I feel like. You should just ask them to, like, do a… like a WeWork. You should ask them just to put up a desk there.
451 00:51:56.470 ⇒ 00:51:57.959 Katherine Bayless: They have offered, yeah.
452 00:51:57.960 ⇒ 00:52:03.960 Uttam Kumaran: Great, like, ASMR, you know, just, like, or, like, whatever, like, ambient sort of movement, showroom.
453 00:52:03.960 ⇒ 00:52:09.149 Katherine Bayless: Yeah, no, I, I could definitely make an office out of this place.
454 00:52:09.150 ⇒ 00:52:17.750 Uttam Kumaran: Okay, awesome. So, thank you guys so much for everything. I’ll hop and get into the game for SDG, but, definitely let’s Talk Monday. Totally.
455 00:52:17.750 ⇒ 00:52:21.889 Katherine Bayless: You guys access to stuff in the meantime, and then, yeah, like, let’s tear this up.
456 00:52:21.890 ⇒ 00:52:33.330 Uttam Kumaran: Yeah, and then if you… anything comes up, and you’re like, I just want to get that to them, I’ll start Slack today. Good. Just livestream thoughts, like, don’t worry at all, we’ll get everything organized and come prepared, so…
457 00:52:33.330 ⇒ 00:52:34.790 Katherine Bayless: I will, I will take you up on that.
458 00:52:34.790 ⇒ 00:52:35.770 Uttam Kumaran: Yeah, yeah, definitely.
459 00:52:36.660 ⇒ 00:52:39.399 Uttam Kumaran: Okay, awesome.
460 00:52:39.400 ⇒ 00:52:40.599 Katherine Bayless: Thank you. Have a great week.
461 00:52:40.890 ⇒ 00:52:42.439 Uttam Kumaran: Yeah, thank you. Bye.