Meeting Title: Brainforge x Default: Roadmap Regroup Date: 2026-05-06 Meeting participants: Scratchpad Notetaker, Caitlyn Vaughn, Uttam Kumaran, Nandika Jhunjhunwala
WEBVTT
1 00:00:52.430 ⇒ 00:00:53.580 Uttam Kumaran: Hey!
2 00:00:53.750 ⇒ 00:00:55.849 Caitlyn Vaughn: Hello! How’s it going?
3 00:00:55.850 ⇒ 00:00:57.470 Uttam Kumaran: Good! How are you?
4 00:00:57.930 ⇒ 00:00:59.459 Caitlyn Vaughn: It’s so early.
5 00:00:59.500 ⇒ 00:01:04.759 Uttam Kumaran: I’m, like, up at, like, 7.30, like…
6 00:01:04.760 ⇒ 00:01:07.700 Caitlyn Vaughn: Yeah, same, but it still feels early.
7 00:01:07.700 ⇒ 00:01:10.430 Uttam Kumaran: Yeah, it does, and it’s, like, gonna rain again, I guess.
8 00:01:10.430 ⇒ 00:01:17.069 Caitlyn Vaughn: I know, for the next week, I’m so incredibly over it, like, especially with the horses.
9 00:01:17.070 ⇒ 00:01:17.760 Uttam Kumaran: Yeah.
10 00:01:17.760 ⇒ 00:01:20.970 Caitlyn Vaughn: You know, they just, like, roll in the mud.
11 00:01:20.970 ⇒ 00:01:22.009 Uttam Kumaran: And then… Oh.
12 00:01:22.010 ⇒ 00:01:27.299 Caitlyn Vaughn: You have to, like, de-crust them every time you ride, and it takes, like, 45 minutes.
13 00:01:27.300 ⇒ 00:01:33.150 Uttam Kumaran: I feel ha- I have a lot of plants, and I just am like, usually it’s such a struggle for them in the summer.
14 00:01:33.150 ⇒ 00:01:34.250 Caitlyn Vaughn: Yeah.
15 00:01:34.250 ⇒ 00:01:41.909 Uttam Kumaran: So I’m like, okay, they’re getting some relief in the lake. Maybe we’ll actually get to go on Lake Travis, like, in a half-decent state.
16 00:01:41.910 ⇒ 00:01:42.790 Caitlyn Vaughn: At some point.
17 00:01:42.790 ⇒ 00:01:44.590 Uttam Kumaran: in my life. I know.
18 00:01:44.590 ⇒ 00:01:48.300 Caitlyn Vaughn: That’s so true. It’s full, like, 85% right now.
19 00:01:48.880 ⇒ 00:01:49.660 Uttam Kumaran: Really?
20 00:01:49.660 ⇒ 00:01:51.510 Caitlyn Vaughn: Yeah, it’s, like, super full.
21 00:01:51.510 ⇒ 00:01:57.370 Uttam Kumaran: Oh, okay, so it’ll be a good summer, because I’ve been, and it’s, like, looks so, like, nasty in there.
22 00:01:57.370 ⇒ 00:02:09.249 Caitlyn Vaughn: Yeah, yeah. I know we… it was, like, down to 35%, I think, 2 years ago, and then last year, it did fill up a little bit, and then this year, it’s just, like, it’s full now.
23 00:02:09.509 ⇒ 00:02:11.389 Uttam Kumaran: Yeah, yeah. It looks good.
24 00:02:11.390 ⇒ 00:02:12.790 Caitlyn Vaughn: Yeah, yeah, yeah.
25 00:02:13.950 ⇒ 00:02:20.989 Caitlyn Vaughn: Okay, so I think Namika will be late, which is fine. I’m feeling better about it.
26 00:02:20.990 ⇒ 00:02:21.420 Uttam Kumaran: Okay.
27 00:02:21.420 ⇒ 00:02:31.330 Caitlyn Vaughn: Greg’s last message, but I just want to make sure that we’re, like, on the same page, so that we’re not spending, like, two weeks, you know, going towards the wrong goal. But…
28 00:02:31.370 ⇒ 00:02:43.720 Caitlyn Vaughn: at a high level, I mean, the DBT work and everything makes a ton of sense. Greg had sent me, like, the DBT strategy that you guys have, which makes a ton of sense, I think that looks good.
29 00:02:44.130 ⇒ 00:02:55.419 Caitlyn Vaughn: the only… like, the major thing I think that we were seeing in the work that was done that needs to be changed is the, like… what is it called, like, RPT tables or something?
30 00:02:55.420 ⇒ 00:02:56.440 Uttam Kumaran: Poor tables, yeah.
31 00:02:56.440 ⇒ 00:02:58.449 Caitlyn Vaughn: Yeah, they’re, like, pre-aggregated tables.
32 00:02:58.450 ⇒ 00:03:12.480 Uttam Kumaran: Yes. So that’s what I basically told… told them, and that’s what they’re doing. It’s like, they’re gonna leave just, like, pre… yeah. So it’s like, yeah, they’re gonna leave no non-aggregated, like, single date spine every.
33 00:03:12.480 ⇒ 00:03:12.840 Caitlyn Vaughn: Sure.
34 00:03:13.080 ⇒ 00:03:19.099 Uttam Kumaran: Yeah. And then I think the second thing they’re basically working on is, like, how much they can move into Omni.
35 00:03:19.100 ⇒ 00:03:19.560 Caitlyn Vaughn: The difficulty.
36 00:03:19.560 ⇒ 00:03:32.429 Uttam Kumaran: is, like, with those new Omni topics, the structure is different, and it’s… you can’t, like… you can’t just, like, point the bar chart to, like, the… the new thing, you know?
37 00:03:32.430 ⇒ 00:03:36.170 Caitlyn Vaughn: Yeah, I know it’s not that easy. Also, will you remove the scratch pad?
38 00:03:37.000 ⇒ 00:03:39.500 Uttam Kumaran: Oh, yeah. It’s not mine. Oh.
39 00:03:39.500 ⇒ 00:03:40.420 Caitlyn Vaughn: It’s mine, yeah.
40 00:03:40.420 ⇒ 00:03:41.450 Uttam Kumaran: Okay, okay, okay.
41 00:03:41.580 ⇒ 00:03:43.790 Caitlyn Vaughn: Yeah, yeah. And that’ll give you the tea.
42 00:03:43.790 ⇒ 00:03:45.040 Uttam Kumaran: Okay, yeah, give back.
43 00:03:45.040 ⇒ 00:03:45.540 Caitlyn Vaughn: brother.
44 00:03:45.540 ⇒ 00:03:50.780 Uttam Kumaran: I was gonna call you about that this month, because I… I can sense some tea, but…
45 00:03:50.780 ⇒ 00:03:51.300 Caitlyn Vaughn: Yeah.
46 00:03:51.300 ⇒ 00:03:52.560 Uttam Kumaran: I don’t know what… yeah.
47 00:03:52.560 ⇒ 00:03:54.920 Caitlyn Vaughn: Honestly, I might need to tell you, like, later.
48 00:03:54.920 ⇒ 00:03:57.540 Uttam Kumaran: Okay, okay, we… I can call… I can call you today.
49 00:03:57.540 ⇒ 00:03:59.119 Caitlyn Vaughn: Yeah, just cause Nanik also, like.
50 00:03:59.120 ⇒ 00:04:00.210 Uttam Kumaran: Sure, sure, sure, sure.
51 00:04:00.210 ⇒ 00:04:04.580 Caitlyn Vaughn: It’s so bad. Okay. I’m, like, I’m, like, about to quit.
52 00:04:04.580 ⇒ 00:04:09.269 Uttam Kumaran: Okay, I’ll call you, I can call you, I’ll call you this week. Are you in Austin this whole week?
53 00:04:09.270 ⇒ 00:04:14.010 Caitlyn Vaughn: I’m in Austin, yeah. I leave next week for San Diego, but I’m in Austin this week.
54 00:04:14.010 ⇒ 00:04:15.279 Uttam Kumaran: Are you, like, busy?
55 00:04:15.730 ⇒ 00:04:17.890 Caitlyn Vaughn: No, I’m not fucking busy, I’m not too.
56 00:04:17.899 ⇒ 00:04:25.149 Uttam Kumaran: Okay, sorry, that’s… I didn’t mean it like that. I was just like… Okay, man, I’ll… I’ll call you today, but…
57 00:04:25.150 ⇒ 00:04:27.270 Caitlyn Vaughn: Yeah, yeah, yeah, call me, call me after this meeting, and I’ll…
58 00:04:27.270 ⇒ 00:04:27.990 Uttam Kumaran: Okay, okay.
59 00:04:27.990 ⇒ 00:04:48.020 Caitlyn Vaughn: yeah, I’m fucking over it anyways. But, yeah, this project, I mean, I feel like it’s going okay. Like, I think the data modeling was probably not done that great, but the true result on the default side is, like, it’s turned into this huge, hot political mess of, like, everyone having a fucking say in it, and it’s…
60 00:04:48.020 ⇒ 00:04:52.030 Uttam Kumaran: Because it’s like, is it… and sorry to interrupt, is it mainly because, like.
61 00:04:52.160 ⇒ 00:04:55.420 Uttam Kumaran: You can now see everything, like, pretty clearly, or on one.
62 00:04:55.420 ⇒ 00:04:56.050 Caitlyn Vaughn: Yeah.
63 00:04:56.050 ⇒ 00:04:56.780 Uttam Kumaran: Okay.
64 00:04:56.780 ⇒ 00:04:58.489 Caitlyn Vaughn: Yeah, I mean, it’s like…
65 00:04:58.670 ⇒ 00:05:15.159 Caitlyn Vaughn: like, Nandica has kind of, like, dug into this always, but I’ve, like, just started teaching myself. I think, actually, let me back up. I think the main reason is because what people can see on the team without having any knowledge of data engineering is…
66 00:05:15.500 ⇒ 00:05:30.529 Caitlyn Vaughn: the Omni outputs, the dashboards, the blobby, you know, the fact that, like, no one can ask Blobby a question and get an accurate response, like, this far into the project is not great. So…
67 00:05:31.020 ⇒ 00:05:45.520 Caitlyn Vaughn: I think at this point, like, then going back and digging into why, it’s, like, all stuff that Nonica has been saying for a long time, but I also just, like, didn’t have enough context to be able to, like, look at it and be like, oh yeah, that is not…
68 00:05:45.790 ⇒ 00:05:50.090 Caitlyn Vaughn: like, based on the things that we want to do, this isn’t gonna work, you know?
69 00:05:50.380 ⇒ 00:05:50.900 Uttam Kumaran: Yeah.
70 00:05:51.360 ⇒ 00:05:55.719 Uttam Kumaran: And part of it is, like, we… we kind of skipped Go and went straight to the dashboards.
71 00:05:56.430 ⇒ 00:06:04.819 Uttam Kumaran: part of it was, like, because that was the core outputs, and we didn’t have… we didn’t get a lot of feedback on the modeling strap, because it was, like, just so.
72 00:06:04.820 ⇒ 00:06:05.140 Caitlyn Vaughn: Yep.
73 00:06:05.140 ⇒ 00:06:06.529 Uttam Kumaran: So the dashboards that…
74 00:06:07.390 ⇒ 00:06:09.720 Uttam Kumaran: Basically, solve their immediate problem.
75 00:06:09.720 ⇒ 00:06:10.150 Caitlyn Vaughn: Yeah.
76 00:06:10.150 ⇒ 00:06:24.700 Uttam Kumaran: And so, this is, like, it’s fairly common to have this, like, rework. Yeah. But, like, it’s, like, moving things into one semantic, it’s, like, this is, like, pretty common once we see people understand the layout.
77 00:06:24.700 ⇒ 00:06:25.200 Caitlyn Vaughn: Yeah.
78 00:06:25.200 ⇒ 00:06:27.680 Uttam Kumaran: But if this takes… this takes time, you know?
79 00:06:27.680 ⇒ 00:06:28.580 Caitlyn Vaughn: Yeah, totally.
80 00:06:28.580 ⇒ 00:06:38.639 Uttam Kumaran: Yeah, so… and Omni, you have, like, the dashboard product, you also have the Blobby product. They kind of, like, the work you do for the dashboard doesn’t necessarily help
81 00:06:38.810 ⇒ 00:06:45.220 Uttam Kumaran: blobby, so it’s, like, another set of, like, all of the AI and the topic building that has to go into that.
82 00:06:45.220 ⇒ 00:06:45.700 Caitlyn Vaughn: Yeah.
83 00:06:45.700 ⇒ 00:06:48.089 Uttam Kumaran: And then we have the ingestion and all the modeling.
84 00:06:48.900 ⇒ 00:06:50.300 Uttam Kumaran: Continue to run, right?
85 00:06:50.300 ⇒ 00:07:02.709 Caitlyn Vaughn: Yeah, totally. Yeah, as I’ve been looking at, like, actually rebuilding this myself, I understand that this is a lot of work. Yeah. But I also, like, I need this to be resolved to a state that we can, like, at least.
86 00:07:02.710 ⇒ 00:07:03.080 Uttam Kumaran: Sure.
87 00:07:03.080 ⇒ 00:07:15.630 Caitlyn Vaughn: lobby and, like, use it so that it’s not as much of a dumpster fire as it is being perceived internally. Okay. So as much as you can help me to, like, get it fixed, I would appreciate it.
88 00:07:16.350 ⇒ 00:07:20.620 Uttam Kumaran: What do you… so then, I basically told them, like, don’t do anything on the dashboard side.
89 00:07:20.620 ⇒ 00:07:21.050 Caitlyn Vaughn: Yeah.
90 00:07:21.050 ⇒ 00:07:27.910 Uttam Kumaran: what I’ll mention is, like, get all the topics together, and I’ll have Audvate basically try and just stress-test Blobby as much as possible.
91 00:07:27.910 ⇒ 00:07:28.300 Caitlyn Vaughn: Yeah.
92 00:07:28.300 ⇒ 00:07:43.289 Uttam Kumaran: I think part of this is, like, we… we have to limit it to just the topics that are, like, the… whatever the final set of topics are. But also, like, I think, Nandika, you can start to… you can even restrict further, like, if we’re not happy with…
93 00:07:43.420 ⇒ 00:07:47.979 Uttam Kumaran: it answering on certain topics, we should just restrict the access until you guys are.
94 00:07:48.670 ⇒ 00:07:57.530 Uttam Kumaran: I mean, part of this is, like, we released the dashboards for QA, but then I know what happens in startups is, like, okay, now I have this, like.
95 00:07:57.760 ⇒ 00:08:00.499 Uttam Kumaran: screenshot this thing, screenshot, like, so…
96 00:08:00.500 ⇒ 00:08:00.840 Caitlyn Vaughn: Yeah.
97 00:08:00.840 ⇒ 00:08:03.869 Uttam Kumaran: Some of it is, like, unavoidable in startup world, like, it just…
98 00:08:04.330 ⇒ 00:08:12.329 Uttam Kumaran: people finally get their hands on something, and, like, we have clients where we’re, like, working on Omni for 6 months, they don’t even release it to, like.
99 00:08:12.560 ⇒ 00:08:13.740 Uttam Kumaran: internally.
100 00:08:13.900 ⇒ 00:08:14.690 Caitlyn Vaughn: Yeah.
101 00:08:14.900 ⇒ 00:08:15.270 Uttam Kumaran: Until.
102 00:08:15.270 ⇒ 00:08:16.429 Caitlyn Vaughn: It’s fully done.
103 00:08:16.430 ⇒ 00:08:17.709 Uttam Kumaran: Until it’s, like, fully gone, so…
104 00:08:17.710 ⇒ 00:08:18.980 Caitlyn Vaughn: Yeah, yeah, yeah.
105 00:08:19.360 ⇒ 00:08:38.779 Uttam Kumaran: it’s… this is just what I know, like, happens, so that’s fine. So I think this week, what basically Greg said, and I think, Nandika, and for you and Caitlin, we can do as many more, like, dbtree trainings, or, like, Omni sort of things, but my big thing is, like, I want to make sure we have those
106 00:08:38.780 ⇒ 00:08:41.509 Uttam Kumaran: Like, non-aggregated…
107 00:08:41.950 ⇒ 00:08:54.779 Uttam Kumaran: models basically ready, and that the topics are built on top of them, so that you can pretty cleanly just build the dashboards ahead of that. There hasn’t been any issues with, like.
108 00:08:55.070 ⇒ 00:08:59.689 Uttam Kumaran: on the ingestion or the, like, dbt running side.
109 00:08:59.690 ⇒ 00:09:00.139 Caitlyn Vaughn: like that.
110 00:09:00.140 ⇒ 00:09:04.000 Uttam Kumaran: It’s all running in GitHub Actions, so maybe that’s, like, a Thomas thing or whatever, so…
111 00:09:04.000 ⇒ 00:09:04.410 Caitlyn Vaughn: Yeah.
112 00:09:04.410 ⇒ 00:09:08.709 Uttam Kumaran: it’s all there. And then, my… my…
113 00:09:09.140 ⇒ 00:09:13.920 Uttam Kumaran: Ask today is, like, what is good in form of, like, documentation, like.
114 00:09:13.960 ⇒ 00:09:33.219 Uttam Kumaran: Is, like, one big, like, wiki helpful? Is, like, what… we have the data platform documentation, but that’s more of, like, a catalog. Like, what is, like, good if we were to just, like, try to throw everything into, like, something where, hey, I have a question about how was… how is Mother Duck set up, all the way up to, like.
115 00:09:33.690 ⇒ 00:09:36.929 Uttam Kumaran: you know, how we name, like, dbt schemas, things like that.
116 00:09:37.560 ⇒ 00:09:43.149 Caitlyn Vaughn: Yeah, I mean, I think from, like, a high level, probably how each
117 00:09:43.800 ⇒ 00:09:51.479 Caitlyn Vaughn: like, each tool in our stack is thought of, like, how you actually built it out would be helpful, and then…
118 00:09:51.670 ⇒ 00:09:53.200 Caitlyn Vaughn: probably, like.
119 00:09:53.550 ⇒ 00:10:04.720 Caitlyn Vaughn: how each topic was curated, what is included in it. Great. I’m almost thinking, like, things that are helpful for us to read, and then things that are also helpful for us to, like, put in a cloud project and be able to use.
120 00:10:04.720 ⇒ 00:10:12.339 Uttam Kumaran: Exactly, yeah, that’s exactly it. So, like, in case you… in case you just throw in a cloud project, and you’re like, where is this, or what is this?
121 00:10:12.340 ⇒ 00:10:13.070 Caitlyn Vaughn: Exactly.
122 00:10:13.070 ⇒ 00:10:18.930 Uttam Kumaran: to answer. I think, like, what I’ll try to deliver is just, like, Basically, like, one big…
123 00:10:19.090 ⇒ 00:10:21.110 Uttam Kumaran: Maybe, like, one big markdown.
124 00:10:21.110 ⇒ 00:10:21.510 Caitlyn Vaughn: Yeah.
125 00:10:21.510 ⇒ 00:10:23.770 Uttam Kumaran: Maybe we’ll even throw that into the repo.
126 00:10:24.720 ⇒ 00:10:25.820 Caitlyn Vaughn: Okay. So that’s, like, sitting…
127 00:10:25.880 ⇒ 00:10:36.999 Uttam Kumaran: there, and I’ll have, so we’ll document everything from the sources to, like, kind of, like, each of the core modeling layer.
128 00:10:37.620 ⇒ 00:10:50.530 Uttam Kumaran: There’ll be some stuff about how we name things, like the scheduling of the jobs, but then, more importantly, Oddvate on our side will sort of go topic by topic. What is this? What tables are in here?
129 00:10:50.580 ⇒ 00:11:00.889 Uttam Kumaran: And then any, like, basically, like, caveats. And then we have documentation on each of the dashboards that I’ll just, like, slot in there, and then maybe just, like.
130 00:11:01.640 ⇒ 00:11:07.940 Uttam Kumaran: archive that Google Doc, basically, and, like, just create, like, one concise File.
131 00:11:08.230 ⇒ 00:11:09.889 Caitlyn Vaughn: Okay. That works.
132 00:11:09.890 ⇒ 00:11:12.359 Nandika Jhunjhunwala: other thing that I would love to see is, like.
133 00:11:13.040 ⇒ 00:11:26.939 Nandika Jhunjhunwala: when we’re talking about, like, a DIM table, or, like, a fact table, or, like, a report table, like, what those tables actually entail, like, I think we looked at previous documentation, and it was a little confusing as to, like, what was exactly in the table, so if.
134 00:11:26.940 ⇒ 00:11:32.710 Uttam Kumaran: Like, if you say, like, in the table, is it more, like, the differences between those types of tables, or, like.
135 00:11:33.090 ⇒ 00:11:33.740 Nandika Jhunjhunwala: Yeah.
136 00:11:33.850 ⇒ 00:11:35.650 Uttam Kumaran: Okay, okay, okay, okay.
137 00:11:35.650 ⇒ 00:11:43.720 Nandika Jhunjhunwala: For example, like, we have, like, I think we had, like, four customer DIM tables, so, like, having, like, demarcation, okay, this is the main customer table, this is, like.
138 00:11:43.720 ⇒ 00:11:44.050 Uttam Kumaran: Okay.
139 00:11:44.050 ⇒ 00:11:51.890 Nandika Jhunjhunwala: DIM table, this is the data in this particular table. I think that would be super helpful, because I think the descriptions were a little bit, like.
140 00:11:51.930 ⇒ 00:12:04.949 Nandika Jhunjhunwala: harder to, like, read or understand, just, like, from the get-go. So we had to, like, dig into it and be like, oh, this is what this is. So if possible at all, to have, like, really clear definitions of what’s in each table would be super helpful for us, like.
141 00:12:04.950 ⇒ 00:12:05.420 Uttam Kumaran: Okay.
142 00:12:05.420 ⇒ 00:12:20.029 Nandika Jhunjhunwala: part of that, like, broader Markdown file. Okay. I’ve also been, like, querying, like, GitHub, like, that repo where all the dbt modeling is done, and I think, like, Claude, from the get-go has, like, great context into, like, what those tables are, and I think that.
143 00:12:20.030 ⇒ 00:12:24.050 Uttam Kumaran: So if you use that with the Mother Duck CLI.
144 00:12:24.050 ⇒ 00:12:24.550 Nandika Jhunjhunwala: Yeah.
145 00:12:24.550 ⇒ 00:12:26.289 Uttam Kumaran: Every time you basically do.
146 00:12:26.290 ⇒ 00:12:26.840 Nandika Jhunjhunwala: stuff, like…
147 00:12:26.980 ⇒ 00:12:34.089 Uttam Kumaran: That’s what I would suggest. Yes. And then Oddvate is also doing a lot with, like, AI, building stuff in Omni.
148 00:12:34.100 ⇒ 00:12:35.329 Nandika Jhunjhunwala: So, like… Yeah.
149 00:12:35.330 ⇒ 00:12:43.629 Uttam Kumaran: I don’t know if he’s explained that, but he’s using, like, the Omni MCP to help him, like, build. Okay, so then that’s, like, also what I was… so, basically, like.
150 00:12:43.630 ⇒ 00:12:47.169 Nandika Jhunjhunwala: That would be also, like, sick to see as, like, a training session.
151 00:12:47.170 ⇒ 00:12:52.139 Uttam Kumaran: Let me ask him to do that, because I would suggest loading all three of those.
152 00:12:52.190 ⇒ 00:13:07.610 Uttam Kumaran: into, like, whatever your cloud session is, and, like, that’s how everybody at Brainforge is, like, kind of doing development and, like, trying to speed things, because previously, like, in a previous world, aka, like, a year ago, you’d literally have to, like, go type in, like, all the YAML, and, like.
153 00:13:08.470 ⇒ 00:13:20.969 Uttam Kumaran: in my career, I literally, like, I had to type everything by hand, all of this, so, like, we’re… we’re a lot further, so I… but it’s also, like, the MCP and their CLI, like, came out, like, 2 weeks ago, so…
154 00:13:20.970 ⇒ 00:13:21.420 Nandika Jhunjhunwala: Yeah.
155 00:13:21.420 ⇒ 00:13:27.309 Uttam Kumaran: We’re actually some of the few people… few people, like, using it, so I would… maybe I’ll ask him to do a session on just, like.
156 00:13:27.520 ⇒ 00:13:34.909 Uttam Kumaran: building with AI and Omni, Omni kind of structures that, so I will… I will ask him to do that.
157 00:13:35.070 ⇒ 00:13:38.140 Nandika Jhunjhunwala: Yeah, I have the same setup, too. Mother Duck CLI, Omni, MCP…
158 00:13:38.710 ⇒ 00:13:39.270 Nandika Jhunjhunwala: You bet.
159 00:13:39.480 ⇒ 00:13:44.529 Uttam Kumaran: It’s just, like, there’s some gotchas with the Omni side, because the stuff is kind of new.
160 00:13:44.730 ⇒ 00:13:50.480 Uttam Kumaran: Like, they literally just released, like, some of it, like, within the last, like, month, so… Okay.
161 00:13:51.370 ⇒ 00:13:57.400 Caitlyn Vaughn: Nautica, do you have that doc? I can’t find it. The one that Demi shared of the naming?
162 00:13:57.610 ⇒ 00:14:00.049 Nandika Jhunjhunwala: Yes, let me… let me send it to you.
163 00:14:00.050 ⇒ 00:14:01.750 Caitlyn Vaughn: Cool, thank you.
164 00:14:03.420 ⇒ 00:14:06.989 Caitlyn Vaughn: It’s interesting to look at this…
165 00:14:07.370 ⇒ 00:14:10.110 Caitlyn Vaughn: Doc, Utom, I don’t know if you saw this.
166 00:14:10.260 ⇒ 00:14:13.630 Caitlyn Vaughn: Okay, wait, so, backing up,
167 00:14:14.470 ⇒ 00:14:27.659 Caitlyn Vaughn: part of the reason why I asked for the… for this meeting is because the message that I got from Greg said, hey, we heard you, we’re gonna move everything to Omni, and that scared me, because that’s Omni.
168 00:14:27.660 ⇒ 00:14:28.040 Uttam Kumaran: Yeah.
169 00:14:28.040 ⇒ 00:14:29.789 Caitlyn Vaughn: Not exactly what is going on.
170 00:14:29.790 ⇒ 00:14:30.659 Uttam Kumaran: Yeah, yeah.
171 00:14:30.660 ⇒ 00:14:41.520 Caitlyn Vaughn: I know he’s, like, not fully technical and, like, probably doesn’t have a lot of context, he’s more of, like, project managing this, but I just want to make sure we’re not… we’re still using dbt, like, we’re still modeling there.
172 00:14:41.520 ⇒ 00:14:51.830 Uttam Kumaran: Yeah, more of the everything was, like, the core logic, but that’s what I wanted to clarify, because that’s what the original doc was like, okay, move the semantic layer up.
173 00:14:52.360 ⇒ 00:15:05.530 Uttam Kumaran: But there’s… semantic layer is kind of a, nebulous, like, term… like, it can mean a lot of things. So I think team is pretty aligned on, like, the core stuff is, like, move to these, like, pre-aggregated tables.
174 00:15:05.560 ⇒ 00:15:17.920 Uttam Kumaran: And then try to land the core, like, case whens, or core joins, all happening in Omni. Okay. Not like… you actually can basically do…
175 00:15:17.920 ⇒ 00:15:26.180 Uttam Kumaran: most of, like, what you can do in dbt in Omni, but then you lose, like… you lose a lot of, like… yeah, Omni becomes the choke point, so…
176 00:15:26.710 ⇒ 00:15:29.050 Uttam Kumaran: not, it’s not, it’s not everything. It’s.
177 00:15:29.050 ⇒ 00:15:29.600 Caitlyn Vaughn: Yeah.
178 00:15:29.600 ⇒ 00:15:31.040 Uttam Kumaran: Yeah, just the core semantic stuff.
179 00:15:31.040 ⇒ 00:15:35.990 Caitlyn Vaughn: Like, the semantic stuff. Okay, and you mean, like, moving away from the pre-aggregated stuff, right?
180 00:15:35.990 ⇒ 00:15:41.510 Uttam Kumaran: Yeah, so, like, previously, we literally pre-aggregated and then just pushed that to Omni to then deliver the dashboard.
181 00:15:41.510 ⇒ 00:15:41.850 Caitlyn Vaughn: Yeah.
182 00:15:41.850 ⇒ 00:15:51.249 Uttam Kumaran: like, if you think about it, like, we have the raw, intermediate is all the, like, pre-aggregated, and then the reports. We’re gonna kind of, like, push that up, basically.
183 00:15:51.250 ⇒ 00:15:51.740 Caitlyn Vaughn: Okay.
184 00:15:51.740 ⇒ 00:16:11.259 Uttam Kumaran: So, like, the kind of, like, the tables where we’re just… all of our opportunities, all of, like, our contacts, we’re gonna make those available all the way into Omni, and the report tables, we pre-aggregated to just support the dashboard. Instead, we’re gonna remove that and just…
185 00:16:11.370 ⇒ 00:16:28.950 Uttam Kumaran: re… like, or we’re gonna… I mean, we’re gonna keep whatever that… so the dashboards are live, but then the pre-aggregated reports will be available as… within topics. So, it’s kind of like pushing, like, the middle layer up, which a lot of the work has actually been done, it’s just we have to, like, rename stuff.
186 00:16:29.820 ⇒ 00:16:33.860 Uttam Kumaran: Consolidate some joins, and then basically push the layer up.
187 00:16:34.070 ⇒ 00:16:41.510 Caitlyn Vaughn: Okay, great. That sounds good. And then, we also… I don’t think we have many measures, I mean, mostly because it was pre-aggregated.
188 00:16:41.510 ⇒ 00:16:41.960 Uttam Kumaran: Yeah.
189 00:16:41.960 ⇒ 00:16:44.449 Caitlyn Vaughn: So, those measures will be in Omni, right?
190 00:16:44.450 ⇒ 00:16:45.359 Uttam Kumaran: Yes, yes, yeah.
191 00:16:45.360 ⇒ 00:16:56.190 Caitlyn Vaughn: Okay. And then the AI context stuff, which is super important for Blobby. Most of it was done, there was, like, I wanna say maybe 25% of it that still needed context, and also
192 00:16:56.610 ⇒ 00:16:58.360 Caitlyn Vaughn: Like, a lot of metadata or anything.
193 00:16:58.360 ⇒ 00:17:16.669 Uttam Kumaran: So that’s actually something I think, Nandika, with AI, with the Claude setup, it would be pretty easy, and, like, that’s what I would suggest, because it’s so much to write. Like, honestly, you guys can probably just throw a lot of internal docs, yeah, in there, and then basically supplement it. And then the other thing I’ll have Advait share is, like.
194 00:17:16.780 ⇒ 00:17:23.299 Uttam Kumaran: we… we… we do this, like, loop-based, like, testing for the… for Blobby, where we, like, try to…
195 00:17:23.319 ⇒ 00:17:37.559 Uttam Kumaran: kind of, like, have a set of questions, we, like, loop and iterate through it, and, like, that was our plan, sort of, to kind of, like, nail, like, okay, we have, like, 50 common questions, let’s loop through, and so I’ll have him sort of share that process a little bit.
196 00:17:38.250 ⇒ 00:17:43.279 Uttam Kumaran: on how to, like, iterate on Blobby, basically, because it’s a little bit different than the dashboards.
197 00:17:43.430 ⇒ 00:17:44.379 Caitlyn Vaughn: Okay, that’s good.
198 00:17:44.380 ⇒ 00:17:54.769 Nandika Jhunjhunwala: The other question I have is, like, during the Omni session, the person from Omni, she shared, like, this, like, really interesting, like, global relationship join…
199 00:17:55.040 ⇒ 00:17:55.800 Uttam Kumaran: Yeah.
200 00:17:56.100 ⇒ 00:17:59.739 Nandika Jhunjhunwala: feature, I think that would be super cool, like, even within a topic.
201 00:17:59.740 ⇒ 00:18:22.739 Nandika Jhunjhunwala: or if any possible joins across topics, just having those defined globally, I think would be, like, super interesting. So then when you’re referring those, like, joins, like, from a base view within a topic in Omni, you just, like, tell it, like, this is another table that you can join against, and it just, like, resolves that relationship globally. So I think that can be super powerful for us, because, like, we basically are…
202 00:18:22.910 ⇒ 00:18:26.399 Nandika Jhunjhunwala: Optimizing for, like, interconnected data with the warehouse.
203 00:18:26.400 ⇒ 00:18:27.090 Uttam Kumaran: Okay.
204 00:18:27.090 ⇒ 00:18:34.200 Nandika Jhunjhunwala: So if possible, for, you know, for you guys to, like, do some preliminary, like, join…
205 00:18:34.290 ⇒ 00:18:35.720 Uttam Kumaran: Yeah, exactly, okay, great.
206 00:18:35.720 ⇒ 00:18:37.910 Nandika Jhunjhunwala: That would be… that would be great, if possible.
207 00:18:38.810 ⇒ 00:18:39.470 Uttam Kumaran: Okay.
208 00:18:41.180 ⇒ 00:18:48.290 Caitlyn Vaughn: Cool. And then this is that doc that Demi shared with us, just so I can show you. I don’t know if you saw this, Utom.
209 00:18:49.110 ⇒ 00:18:59.890 Caitlyn Vaughn: This is the list of all of the different models, plus, like, an explanation as to what they are. Yeah. So, obviously, some of these are, like, source-specific.
210 00:18:59.890 ⇒ 00:19:00.470 Uttam Kumaran: Yes.
211 00:19:00.470 ⇒ 00:19:04.460 Caitlyn Vaughn: which we want to move away from. But then, also, just, like, some of the.
212 00:19:04.460 ⇒ 00:19:12.159 Uttam Kumaran: So, like, one piece on the source-specific models is, like, we will create a DIM customer per source, and then ladder it up.
213 00:19:12.500 ⇒ 00:19:13.340 Uttam Kumaran: So…
214 00:19:13.340 ⇒ 00:19:13.850 Caitlyn Vaughn: Okay.
215 00:19:13.850 ⇒ 00:19:19.320 Uttam Kumaran: I guess my point is, like, it is… it will be there, but whether it’s surfaced
216 00:19:19.570 ⇒ 00:19:20.990 Uttam Kumaran: in Omni is, like.
217 00:19:20.990 ⇒ 00:19:21.490 Caitlyn Vaughn: Yeah.
218 00:19:21.490 ⇒ 00:19:23.660 Uttam Kumaran: question. Okay. You kind of see what I mean?
219 00:19:23.660 ⇒ 00:19:24.350 Caitlyn Vaughn: It’s a model.
220 00:19:24.350 ⇒ 00:19:29.289 Uttam Kumaran: choice to, like, go source by source. We basically create the same data model, and then we
221 00:19:30.280 ⇒ 00:19:33.009 Uttam Kumaran: Like, the same rain? Yeah, exactly.
222 00:19:33.010 ⇒ 00:19:38.210 Caitlyn Vaughn: Okay, interesting. Yeah, that’s fine. Okay, that makes a ton of sense, then.
223 00:19:39.070 ⇒ 00:19:45.279 Caitlyn Vaughn: And then for the explanations, I think some of them were still pretty confusing, okay.
224 00:19:45.620 ⇒ 00:19:50.370 Caitlyn Vaughn: Like, I’m trying to think of an example, one playing customer support per day…
225 00:19:54.830 ⇒ 00:20:09.989 Caitlyn Vaughn: I don’t know. Let’s look at the end of this. One integration times team with count of workflow using it. Okay, so a fact workflow count by integration table. So one integration per team with count of workflow using it.
226 00:20:09.990 ⇒ 00:20:12.570 Uttam Kumaran: There’s no day… there’s no day grain on this, basically, Ben.
227 00:20:13.170 ⇒ 00:20:14.430 Caitlyn Vaughn: Yeah, but…
228 00:20:14.810 ⇒ 00:20:18.590 Uttam Kumaran: This is, like, inter… integration, team… And then…
229 00:20:18.590 ⇒ 00:20:19.780 Caitlyn Vaughn: What is team?
230 00:20:20.060 ⇒ 00:20:23.620 Uttam Kumaran: The default, customer, like the…
231 00:20:23.620 ⇒ 00:20:23.990 Caitlyn Vaughn: Okay.
232 00:20:23.990 ⇒ 00:20:26.410 Uttam Kumaran: But, okay, fair, we should use, like.
233 00:20:27.400 ⇒ 00:20:29.270 Uttam Kumaran: We should just say customer team.
234 00:20:29.560 ⇒ 00:20:40.229 Caitlyn Vaughn: Yeah, like, I guess that makes sense. We don’t use the word team, so to me that is, like, a little bit confusing, because it’s just such a… such a short explanation, right?
235 00:20:40.230 ⇒ 00:20:40.650 Uttam Kumaran: Yeah.
236 00:20:40.650 ⇒ 00:20:42.339 Caitlyn Vaughn: was longer or had more context than I thought.
237 00:20:42.340 ⇒ 00:20:53.100 Uttam Kumaran: Kind of a couple things we can do here is, like, we could do one, we can just say, like, here’s, like, the business case for the table, here’s, like, an example question that you would answer.
238 00:20:53.650 ⇒ 00:20:59.820 Uttam Kumaran: I mean, ultimately, what you’re gonna see is, like, we should just move to, like, fact workflow.
239 00:20:59.820 ⇒ 00:21:00.250 Caitlyn Vaughn: Yeah.
240 00:21:00.250 ⇒ 00:21:01.449 Uttam Kumaran: And then it’s like…
241 00:21:01.670 ⇒ 00:21:12.209 Uttam Kumaran: has a lot of these in it. So part of the reason for these is, yeah, we just moved fast on the modeling to be able to support the dashboard, so you see these, like, fragmentation tables.
242 00:21:12.700 ⇒ 00:21:17.090 Uttam Kumaran: Like, so we can just move to basically collapse as much of this into, like.
243 00:21:17.470 ⇒ 00:21:34.170 Uttam Kumaran: What will happen, though, is these tables sort of just get very long. Yeah. And so we’ll have, like, the day, the integration, maybe just the team ID, so that you can join back to, like, dim teams, or whatever, and then all of the measures, and then that we push up.
244 00:21:34.390 ⇒ 00:21:35.690 Caitlyn Vaughn: Okay.
245 00:21:35.690 ⇒ 00:21:37.689 Uttam Kumaran: Those little tables are actually built
246 00:21:37.830 ⇒ 00:21:40.219 Uttam Kumaran: On, like, a layer of that.
247 00:21:40.220 ⇒ 00:21:40.810 Caitlyn Vaughn: Yeah, yeah, yeah.
248 00:21:40.810 ⇒ 00:21:49.560 Uttam Kumaran: So, that’s what we’ll… we’re basically just, like, moving… promoting that up, all the way into Omni. So not just into Martz, but, like, all the way to the top.
249 00:21:49.560 ⇒ 00:21:49.930 Caitlyn Vaughn: Yeah.
250 00:21:49.930 ⇒ 00:22:01.029 Uttam Kumaran: If you take that example of, like, fact, like, workflow usage, and let’s say it’s, like, day, workflow integration, whatever, in there, let’s say there’s a team ID,
251 00:22:01.280 ⇒ 00:22:10.080 Uttam Kumaran: in Omni, in the topic, we would indicate that as a join. So you can join that to anything that includes the team ID. So that’s, like, kind of the through-line example.
252 00:22:12.380 ⇒ 00:22:21.779 Nandika Jhunjhunwala: And then, like, just to clarify, like, with measures we can define as, like, things we want to aggregate on, so that could be by integration, it could be, like, by time period.
253 00:22:22.150 ⇒ 00:22:22.710 Uttam Kumaran: Yeah.
254 00:22:22.710 ⇒ 00:22:25.040 Nandika Jhunjhunwala: Any other sort of, like, aggregations we want to do, we can.
255 00:22:25.040 ⇒ 00:22:30.079 Uttam Kumaran: So, yeah, everything will exist as a measure, and then you can create metrics on top of that.
256 00:22:30.080 ⇒ 00:22:30.590 Nandika Jhunjhunwala: Absolutely.
257 00:22:30.680 ⇒ 00:22:39.399 Uttam Kumaran: All in Omni, so you won’t have to go back to dbt to create it, so you can create it all in Omni, and then…
258 00:22:39.870 ⇒ 00:22:43.089 Uttam Kumaran: But, like, yeah, the measures will show up.
259 00:22:43.200 ⇒ 00:22:46.589 Uttam Kumaran: all the way up through the top, basically.
260 00:22:47.510 ⇒ 00:22:59.270 Uttam Kumaran: Yeah, so it should be… it should be easier to go from, like, we need this, to, okay, we can do this change, directly combining these measures into some logic, and then test it all within one branch.
261 00:22:59.690 ⇒ 00:23:04.199 Uttam Kumaran: show that that dashboard’s working, and then, like, basically promote it.
262 00:23:04.380 ⇒ 00:23:05.140 Caitlyn Vaughn: Okay.
263 00:23:05.870 ⇒ 00:23:06.510 Uttam Kumaran: Yeah.
264 00:23:07.410 ⇒ 00:23:09.270 Caitlyn Vaughn: Okay, that makes sense.
265 00:23:09.600 ⇒ 00:23:18.459 Caitlyn Vaughn: And then, with my messages with Greg as well, when I was like, okay, we need to, like, redo this, he was like, we have 20 hours left in our contract.
266 00:23:18.460 ⇒ 00:23:18.930 Uttam Kumaran: Yeah.
267 00:23:18.930 ⇒ 00:23:22.920 Caitlyn Vaughn: Is that… I thought we were doing 20 hours per week, is it 10 hours per week?
268 00:23:23.200 ⇒ 00:23:28.320 Uttam Kumaran: We are doing, yeah, we’re doing 10 hours per week.
269 00:23:28.480 ⇒ 00:23:29.310 Caitlyn Vaughn: Oh, really?
270 00:23:30.080 ⇒ 00:23:31.840 Uttam Kumaran: 40 hours a month.
271 00:23:31.840 ⇒ 00:23:33.030 Caitlyn Vaughn: Oh…
272 00:23:33.030 ⇒ 00:23:35.379 Uttam Kumaran: Okay. And so…
273 00:23:35.830 ⇒ 00:23:42.009 Uttam Kumaran: Well, I mean, look, I think ultimately, I want to make sure we land the plane on this, so that’s what I told Greg.
274 00:23:42.020 ⇒ 00:23:59.499 Uttam Kumaran: Greg is also on the hook for making sure that we’re hitting margin, so that’s the sort of back and forth. I think, like, let me take what we said today. I think we have, like, the core pre-aggregated, items that need to make it into Omni all the way up.
275 00:23:59.750 ⇒ 00:24:08.509 Uttam Kumaran: Nandica, we have the, like, Advait doing a demo session. We have also the configuration of the global joins.
276 00:24:08.710 ⇒ 00:24:13.900 Uttam Kumaran: in Omni, and then finally we have, like, giant wiki.
277 00:24:14.220 ⇒ 00:24:19.629 Uttam Kumaran: That’s, like, kind of the core four pieces, so let me deliver that to him and say, like.
278 00:24:19.860 ⇒ 00:24:22.810 Uttam Kumaran: Break this into who’s doing it in hours.
279 00:24:23.310 ⇒ 00:24:27.930 Uttam Kumaran: And then, like, maybe the three of us just, like, have a conversation on how long he’s gonna take.
280 00:24:29.030 ⇒ 00:24:30.550 Uttam Kumaran: How long do you think it’s gonna take?
281 00:24:30.940 ⇒ 00:24:33.300 Caitlyn Vaughn: Yeah, that sounds good.
282 00:24:33.570 ⇒ 00:24:38.999 Caitlyn Vaughn: Cool. Is there anything else that we, like, should clarify on our side, or, like, any questions.
283 00:24:39.000 ⇒ 00:24:49.869 Uttam Kumaran: So, I’m gonna make sure in that doc, too, like, whoever we’ve talked to regarding each of those vendors, like, you have Gallup’s contact, you have the… I’ll give you the mother duck contact…
284 00:24:50.210 ⇒ 00:24:53.840 Uttam Kumaran: You have the Omni contact,
285 00:24:59.100 ⇒ 00:25:06.430 Uttam Kumaran: Yeah, and then also, even, like, I’m also here, and, like, I don’t, like, even after this, like, I can call Brian, and he’ll come, he can be helpful, so, like.
286 00:25:06.430 ⇒ 00:25:06.820 Caitlyn Vaughn: Hmm.
287 00:25:06.820 ⇒ 00:25:08.180 Uttam Kumaran: That’s the app.
288 00:25:08.290 ⇒ 00:25:17.800 Uttam Kumaran: I… we’re gonna be around just to, like, just to make sure that if you guys need anything, so just let me know. I think, like, some of the team is gonna move on to work on.
289 00:25:17.800 ⇒ 00:25:18.150 Caitlyn Vaughn: stuff.
290 00:25:18.150 ⇒ 00:25:24.700 Uttam Kumaran: So… but I’m gonna be on, and again, it’s like, you want Brian to come just do another session just around dbt, so…
291 00:25:25.570 ⇒ 00:25:27.019 Uttam Kumaran: We’re, like, alive, and, like…
292 00:25:27.020 ⇒ 00:25:27.620 Caitlyn Vaughn: Yeah, beer.
293 00:25:27.620 ⇒ 00:25:34.860 Uttam Kumaran: So, I’m not, like, I don’t… I’m not… I’m not, like, super stingy about that, so I just want to make sure we figure this out in the next two weeks.
294 00:25:34.860 ⇒ 00:25:38.050 Caitlyn Vaughn: Yeah, sounds good. Also, Brian was awesome. We really liked him a lot.
295 00:25:38.050 ⇒ 00:25:40.699 Uttam Kumaran: Brian’s really great. So, you know, Brian,
296 00:25:40.820 ⇒ 00:25:43.660 Uttam Kumaran: Brian got me my first job at WeWork, actually.
297 00:25:43.660 ⇒ 00:25:45.799 Caitlyn Vaughn: Yeah, you just work there together, right?
298 00:25:45.800 ⇒ 00:25:48.869 Uttam Kumaran: I cold emailed him when I was a senior in college.
299 00:25:48.870 ⇒ 00:25:49.220 Caitlyn Vaughn: No way.
300 00:25:49.220 ⇒ 00:26:03.040 Uttam Kumaran: some sappy email to ask him for… to tell me about his job in BI, and he called me, and he was like… he called me, we talked, he was like, look, you don’t know anything about business intelligence, but I like you. Let me just, like…
301 00:26:03.040 ⇒ 00:26:14.090 Uttam Kumaran: let me see what we can do, and then, yeah, he got me that job at WeWork, and then, so, we work together there, and then, yeah, he has a kind of a long career doing a bunch of stuff. I would love for him to work for us.
302 00:26:14.090 ⇒ 00:26:19.199 Uttam Kumaran: But he is, he is, just hanging out at Spotify in New York, and .
303 00:26:19.200 ⇒ 00:26:20.530 Caitlyn Vaughn: Oh, he’s at Spotify?
304 00:26:20.530 ⇒ 00:26:23.909 Uttam Kumaran: Yeah, so he’s helped us in and out for a while.
305 00:26:24.260 ⇒ 00:26:31.239 Uttam Kumaran: At the company. Yeah. But I have not been able to get him full-time, because his job is very, very cushy there.
306 00:26:31.240 ⇒ 00:26:32.280 Caitlyn Vaughn: Oh, really?
307 00:26:32.280 ⇒ 00:26:32.820 Uttam Kumaran: Yeah.
308 00:26:32.820 ⇒ 00:26:33.160 Caitlyn Vaughn: Hey, Pat.
309 00:26:33.160 ⇒ 00:26:42.419 Uttam Kumaran: But he’s like, he basically told me, he’s like, AI’s coming for everything. They may let me go. I said, well, dude, you’re great, you should come work here. So, yeah, so me and him are very similar…
310 00:26:42.530 ⇒ 00:26:49.910 Uttam Kumaran: kind of upbringing in data, although he’s a bit older than me, but both of us did a ton of dbt and data work for a long time, but…
311 00:26:49.910 ⇒ 00:26:51.299 Caitlyn Vaughn: Yeah. Yeah, he’s great.
312 00:26:51.300 ⇒ 00:26:52.949 Uttam Kumaran: Patient, and so, yeah.
313 00:26:53.270 ⇒ 00:27:01.239 Caitlyn Vaughn: Amazing. Okay, cool. Well… I think the, like, the most helpful thing that we could do is just…
314 00:27:01.470 ⇒ 00:27:02.970 Caitlyn Vaughn: Like, write down what.
315 00:27:02.970 ⇒ 00:27:04.170 Uttam Kumaran: Yes. The score end result.
316 00:27:04.170 ⇒ 00:27:08.110 Caitlyn Vaughn: is, like, this is what we’re aiming for, so that we have a good sense of, like.
317 00:27:08.290 ⇒ 00:27:19.829 Caitlyn Vaughn: things, you know, ended up in a good place for us. Sure. I did outline, like, I sent a message in our, like, more private chat, the brain… data brain.
318 00:27:19.830 ⇒ 00:27:20.480 Uttam Kumaran: data team?
319 00:27:20.480 ⇒ 00:27:32.329 Caitlyn Vaughn: Yeah, yeah, yeah, that one. With kind of, like, what our requirements are for this project. Some of that, like, the dashboarding stuff, that’s gonna have to happen, you know, after you guys have done the modeling and are probably gone.
320 00:27:32.330 ⇒ 00:27:32.910 Uttam Kumaran: Yeah.
321 00:27:32.910 ⇒ 00:27:37.210 Caitlyn Vaughn: But hopefully we’re in a better place to where that’s a little bit easier to do.
322 00:27:37.210 ⇒ 00:27:51.519 Uttam Kumaran: Yeah. So what… how about what I do is I’m gonna take your bullets, I’m gonna join what’s on my bullets, and I’m just gonna put, like, Brain Forge, default, and let’s just, like, we just iterate back and forth on that. Yeah. Like, for example, like, every topic needs…
323 00:27:51.520 ⇒ 00:28:04.850 Uttam Kumaran: human-readable metadata in context. I’m gonna split that with you, Nautica, basically, if that’s easy to do, but then… so that’s… so I’ll just do that with a couple of items that we had today, and then I’ll just… yeah, Greg will just drive those forward.
324 00:28:04.850 ⇒ 00:28:15.260 Caitlyn Vaughn: Yeah, and that’s also such a good point. We can put hours into this, so if there are things that are, like, low lift, but high hours, like, put that on us, and we will… we’ll run with that.
325 00:28:15.260 ⇒ 00:28:24.009 Uttam Kumaran: Okay, so that’s what I’ll try to do. I basically want to make sure that the wiki is super, super important, getting all the pre-aggregated stuff to the top.
326 00:28:24.010 ⇒ 00:28:24.410 Caitlyn Vaughn: It’s really cool.
327 00:28:24.410 ⇒ 00:28:31.980 Uttam Kumaran: really important. And then I want to do… have Audbe do that, training, Nandica, so that
328 00:28:32.110 ⇒ 00:28:41.030 Uttam Kumaran: you’ll probably, hopefully, just be, like, 20-40% faster on, like, Omni development, so… Perfect.
329 00:28:43.800 ⇒ 00:28:44.810 Caitlyn Vaughn: Awesome.
330 00:28:45.200 ⇒ 00:28:45.580 Uttam Kumaran: Okay.
331 00:28:45.580 ⇒ 00:28:46.160 Caitlyn Vaughn: world.
332 00:28:46.400 ⇒ 00:28:48.050 Caitlyn Vaughn: Thank you, Utam.
333 00:28:48.050 ⇒ 00:28:50.680 Uttam Kumaran: Thank you. Thank you so much. Appreciate y’all. Okay, talk soon.
334 00:28:50.680 ⇒ 00:28:51.800 Caitlyn Vaughn: Alright, talk to you later. Bye.
335 00:28:51.800 ⇒ 00:28:52.340 Uttam Kumaran: Bye.