Meeting Title: Brainforge __ Eden - Kick off meeting Date: 2025-01-09 Meeting participants: Luke Daque, Nicolas Sucari, Uttam Kumaran, Bo Yoon, Robert Tseng
WEBVTT
1 00:00:42.430 ⇒ 00:00:43.660 Uttam Kumaran: No further.
2 00:00:43.810 ⇒ 00:00:51.099 Uttam Kumaran: We offer strategy as a minimum meaning, like we won’t take any deal where we don’t help them
3 00:00:51.190 ⇒ 00:01:16.070 Uttam Kumaran: think about like how we’re going to execute that because we’ve had a bunch of clients who bring us in there like here’s our plans. We just executed a shit plan. So we come in and basically think through like, what do you? What’s your event structure like? What are you tracking? What are your current dashboards? Are you even using data to make decisions. And then we isolate like, it’s probably a couple of problems kind of like, we need to restructure that we need to make like Xy.
4 00:01:24.980 ⇒ 00:01:40.749 Uttam Kumaran: Some people they bring us off because they’re like they think that they’re in for as bad. They just don’t know how some people are like married to it. And I’m like cool, and we’ll try to come up with strategy that works within what you guys have and what you guys can afford to do. The reason why we’re starting to do that is like, sometimes we got thrown in a situation where.
5 00:01:41.100 ⇒ 00:01:53.039 Uttam Kumaran: like we just couldn’t like. They just wanted us to come in like, almost just like, write the code. But like guarantee that we’re gonna succeed and also have some ability to craft.
6 00:03:49.390 ⇒ 00:03:53.020 Uttam Kumaran: Hi, hey?
7 00:03:53.240 ⇒ 00:03:58.279 Uttam Kumaran: I just got just got home.
8 00:03:58.510 ⇒ 00:04:05.300 Uttam Kumaran: Yeah, where were you? I was at ABC home and commercial.
9 00:04:05.610 ⇒ 00:04:11.739 Uttam Kumaran: That’s weird. Why is random? How do they go?
10 00:04:13.000 ⇒ 00:04:14.619 Uttam Kumaran: It was good.
11 00:04:15.392 ⇒ 00:04:18.819 Uttam Kumaran: Yeah. It was actually so super, it was actually amazing.
12 00:04:19.029 ⇒ 00:04:23.160 Uttam Kumaran: You were pitching them, or they’re a current client. We were pitch. We were pitching them
13 00:04:23.550 ⇒ 00:04:26.420 Uttam Kumaran: on 2 AI agents, honor.
14 00:04:26.700 ⇒ 00:04:36.839 Uttam Kumaran: No, no. I just went with Scott honestly, like we’re we’re doing what like I told you we I wanted to kind of do is like, we’re building an AI agent for all their Csr reps.
15 00:04:37.200 ⇒ 00:04:44.649 Uttam Kumaran: yeah. So like, imagine you’re on a call and like a client asks you a question, and you’re like frantically trying to go to a document
16 00:04:45.800 ⇒ 00:04:49.000 Uttam Kumaran: to to find the answer if it stays open.
17 00:04:49.220 ⇒ 00:05:05.857 Uttam Kumaran: Well, let’s say, because your company is maybe a little bit more complicated. You have like a ho! A 100 documents, or like let’s say you service all of Texas for every single pest that could ever involve your house. You probably have probably have a ton of documents.
18 00:05:07.490 ⇒ 00:05:08.650 Uttam Kumaran: so yeah.
19 00:05:10.770 ⇒ 00:05:18.049 Uttam Kumaran: Well, I got the job offer. Oh, nice, hey, Nico, give me one sec. I’ll be right on.
20 00:05:20.960 ⇒ 00:05:22.060 Nicolas Sucari: Yeah, no. Problem.
21 00:06:20.930 ⇒ 00:06:21.820 Robert Tseng: Everyone.
22 00:06:24.270 ⇒ 00:06:26.099 Luke Daque: Hey, guys, how’s it going.
23 00:06:27.350 ⇒ 00:06:28.509 Robert Tseng: Good. How are you guys.
24 00:06:29.870 ⇒ 00:06:31.449 Luke Daque: Doing well, doing well.
25 00:06:32.280 ⇒ 00:06:33.280 Nicolas Sucari: Hey, guys?
26 00:06:40.950 ⇒ 00:06:46.200 Nicolas Sucari: Rob told. Some declined the invite today.
27 00:06:46.420 ⇒ 00:06:55.340 Nicolas Sucari: he said. He’s not gonna gonna be able to join. I don’t know, Robert. If you prefer to move it for tomorrow. I can ask Rob if he’s available same time tomorrow.
28 00:06:58.890 ⇒ 00:07:04.190 Robert Tseng: I mean, I I think we should just we should just meet. We should just use this as our inter internal meeting.
29 00:07:04.190 ⇒ 00:07:05.060 Nicolas Sucari: Yeah, yeah.
30 00:07:05.370 ⇒ 00:07:05.790 Robert Tseng: Yeah.
31 00:07:05.790 ⇒ 00:07:07.919 Nicolas Sucari: Bo is here too. So hi, bo!
32 00:07:08.160 ⇒ 00:07:09.979 Bo Yoon: Hey? How’s it going.
33 00:07:10.310 ⇒ 00:07:11.810 Uttam Kumaran: Hey? Hey? Guys.
34 00:07:14.700 ⇒ 00:07:15.560 Nicolas Sucari: 8 times.
35 00:07:21.440 ⇒ 00:07:24.510 Uttam Kumaran: Yeah, I agree, we can just review everything in flight. Come down.
36 00:07:26.340 ⇒ 00:07:39.230 Robert Tseng: Yeah, if anything, I also want Bo to be seeing what you guys are doing. Because, like, I mean, he, his. His analysis is built on top of the models that we’re kind of moving to. Dbt, so.
37 00:07:40.990 ⇒ 00:07:46.850 Uttam Kumaran: Okay. Maybe, like, we can even start with that or give an overview of where we are.
38 00:07:47.190 ⇒ 00:07:48.949 Uttam Kumaran: Oh, however, you want to take it.
39 00:07:50.417 ⇒ 00:07:59.352 Robert Tseng: Yeah. I mean, why don’t we do like an overview? Just because I don’t know if Bo has seen, like the dB, 2 workflow before.
40 00:08:00.660 ⇒ 00:08:02.110 Robert Tseng: yeah, he’s on.
41 00:08:02.110 ⇒ 00:08:05.850 Bo Yoon: I’m not aware of. I’m not really used to that.
42 00:08:06.480 ⇒ 00:08:09.819 Uttam Kumaran: Yeah, I’m happy to. I can take it and just walk through
43 00:08:10.290 ⇒ 00:08:12.381 Uttam Kumaran: sort of how Dbt works. And
44 00:08:13.430 ⇒ 00:08:16.364 Uttam Kumaran: basically, how doing stuff. And okay, let me just
45 00:08:17.230 ⇒ 00:08:18.519 Bo Yoon: Yeah, that would be great. Thank you.
46 00:08:18.520 ⇒ 00:08:25.979 Uttam Kumaran: So open a bigquery. And then, right after that, I think let’s talk. I don’t know how much if we’ve done any work, Ryan, yet on migrating those
47 00:08:27.940 ⇒ 00:08:31.420 Uttam Kumaran: step over or anything. But we can even talk about what the plan is.
48 00:08:31.750 ⇒ 00:08:34.171 Luke Daque: Yeah, sure, I think I’ve done a couple of
49 00:08:34.580 ⇒ 00:08:36.989 Luke Daque: of the scheduled queries migrating them.
50 00:08:38.010 ⇒ 00:08:38.770 Luke Daque: Yeah.
51 00:08:39.570 ⇒ 00:08:44.639 Robert Tseng: Yeah, we should talk priorities of which ones to migrate over. I know there’s a lot. So
52 00:08:45.298 ⇒ 00:08:47.590 Robert Tseng: definitely anything related to like
53 00:08:48.010 ⇒ 00:08:56.360 Robert Tseng: product. Like, I think there’s 1 or the order details table. Materialized view is probably the the biggest priority one, just since, like
54 00:08:57.020 ⇒ 00:09:06.039 Robert Tseng: that is basically their one big table model for all like relevant order, level details.
55 00:09:06.330 ⇒ 00:09:12.886 Luke Daque: Yeah, I did notice it’s being used by a lot of other queries, and that’s already done, though I mean, I’ve already like
56 00:09:14.000 ⇒ 00:09:16.210 Luke Daque: migrated that in the yeah.
57 00:09:16.600 ⇒ 00:09:17.550 Robert Tseng: Okay. Great.
58 00:09:19.960 ⇒ 00:09:26.989 Uttam Kumaran: Okay, awesome. So we are. I’ll kind of give you a sense of one like how we’re organizing things. So we have
59 00:09:27.200 ⇒ 00:09:30.050 Uttam Kumaran: a notion board with sort of tickets on.
60 00:09:30.557 ⇒ 00:09:46.952 Uttam Kumaran: Basically, all the work that we’re doing. We have a lot of migration tickets that we’re working on that are kind of split between me and Ryan, where we’re basically moving over. The save queries that are here scheduled queries that are here at Dbt, of course.
61 00:09:47.490 ⇒ 00:09:54.829 Uttam Kumaran: for everybody, the schedule query, is basically just a large set of just basically a large
62 00:09:55.200 ⇒ 00:10:03.720 Uttam Kumaran: query that we basically will 1st move over to make sure they’re getting executed by Dbt, and then, second, we’ll start to break this up into
63 00:10:03.900 ⇒ 00:10:07.370 Uttam Kumaran: sort of models as as necessary.
64 00:10:07.958 ⇒ 00:10:14.770 Uttam Kumaran: So to talk about sort of how we’re doing that. We’re using. I’m just going to show
65 00:10:15.830 ⇒ 00:10:45.579 Uttam Kumaran: this we’re using. Dbt, let me just get rid of this. Dbt is basically, you think about it as like an orchestrator, and a development interface for writing sequel. We will be creating several data models that may pull from each other, combine. And so we need somewhere where we can host all that, and then also host the execution of these models on a schedule right? Looking at all the scheduled queries. Looks like we have a range from like every 30 min to like
66 00:10:45.620 ⇒ 00:11:03.083 Uttam Kumaran: once a day, so we’ll make sure that there’s jobs for all that. So to start, we basically have like one deployment right now, which is just a daily production. As you can see this is all just set up yesterday, so nothing is sort of hooked into this right now, but we’ll probably end up having a few different
67 00:11:03.867 ⇒ 00:11:28.810 Uttam Kumaran: production, sort of jobs. One that’s daily, probably one that’s hourly. And then whatever something in between, we also have our development. So if you go to Cloud Ide, you’ll actually see in here all the sort of code that we’re writing. To create. Basically, all the models. To give you a sense of like, what a model is. This is our actually, let me just go to.
68 00:11:29.510 ⇒ 00:11:37.917 Uttam Kumaran: This is an example of the Cac by product. Sequel, if we if we go into actually like where this is
69 00:11:39.260 ⇒ 00:11:56.739 Uttam Kumaran: we basically took this query, I want to be like, basically super specific. Here’s like, at the query, we took this, cleaned it up and basically created this table. This is that, except probably just like formatted away better. And this has all the logic, as you can see, even in this current state. It’s still
70 00:11:56.850 ⇒ 00:12:09.549 Uttam Kumaran: kind of messy. But we’re gonna one. We just wanna make sure that we do a lift and shift where everything gets executed by Dbt, and you’ll actually see that we are executing these in
71 00:12:10.636 ⇒ 00:12:14.204 Uttam Kumaran: in bigquery. And they’re actually landing here in
72 00:12:15.200 ⇒ 00:12:18.630 Uttam Kumaran: dbt march. So there’s probably gonna be Cac.
73 00:12:19.390 ⇒ 00:12:22.869 Uttam Kumaran: If there isn’t already, it’ll probably appear here soon where we have the
74 00:12:23.140 ⇒ 00:12:30.810 Uttam Kumaran: the Cac table. And you can see here, these are all being executed by dbt, right now.
75 00:12:31.307 ⇒ 00:12:42.239 Uttam Kumaran: So that’s basically what we’re doing. We have like 24 ish models that we’re going through. Just basically moving them to here once they’re here. What we’ll do is we’ll start to see if we can
76 00:12:42.340 ⇒ 00:12:58.671 Uttam Kumaran: have you or your team start to use these instead of the locations for the scheduled queries. And we’ll also start to go, basically, business domain by business domain and start to break these down into the relevant
77 00:13:00.140 ⇒ 00:13:02.210 Uttam Kumaran: sort of tables.
78 00:13:02.579 ⇒ 00:13:04.970 Uttam Kumaran: So that’s kind of like where we are on the Dbt side.
79 00:13:06.360 ⇒ 00:13:06.970 Bo Yoon: Yeah.
80 00:13:07.660 ⇒ 00:13:09.785 Uttam Kumaran: I guess I had one question
81 00:13:10.360 ⇒ 00:13:20.890 Uttam Kumaran: or Luke like, are we? I didn’t know we were breaking these down yet, or these just like table, are the intermediate tables, just tables that are getting used in several places.
82 00:13:21.576 ⇒ 00:13:24.780 Luke Daque: I just added those there because those are the
83 00:13:26.960 ⇒ 00:13:35.539 Luke Daque: the views that are already currently in bigquery. They’re not in the scheduled queries, but they are stored as views.
84 00:13:35.970 ⇒ 00:13:36.649 Luke Daque: So I just added
85 00:13:36.650 ⇒ 00:13:41.529 Luke Daque: them there at the moment. We can. We can further break that down further. Once we’re
86 00:13:42.250 ⇒ 00:13:45.250 Luke Daque: gonna refactor and like clean, clean everything up.
87 00:13:45.520 ⇒ 00:13:47.810 Uttam Kumaran: And these are all. These are all views as well.
88 00:13:48.220 ⇒ 00:13:54.110 Luke Daque: Yeah for the staging. Those are like coming from the raw source sources.
89 00:13:54.630 ⇒ 00:13:58.829 Luke Daque: And basically, I just created staging views for them in case, we need to
90 00:13:59.720 ⇒ 00:14:03.480 Luke Daque: like rename stuff or like very basic transformation.
91 00:14:04.750 ⇒ 00:14:07.560 Uttam Kumaran: Okay. But you’re not using these in any models yet. Right?
92 00:14:07.560 ⇒ 00:14:11.300 Luke Daque: Those are being used or the march models based.
93 00:14:11.300 ⇒ 00:14:14.099 Uttam Kumaran: Okay, I just wanna do. I just don’t wanna do
94 00:14:15.310 ⇒ 00:14:17.870 Uttam Kumaran: like, I just really don’t want to do everything all at once.
95 00:14:18.000 ⇒ 00:14:21.309 Uttam Kumaran: I want us to be very careful, because there’s a lot going on.
96 00:14:21.776 ⇒ 00:14:23.989 Uttam Kumaran: I don’t want us to take like.
97 00:14:25.090 ⇒ 00:14:31.539 Uttam Kumaran: I mean, there’s the the core thing. Short term is just to make sure that we’re running everything through. Dbt.
98 00:14:32.100 ⇒ 00:14:32.670 Luke Daque: Right.
99 00:14:32.670 ⇒ 00:14:35.519 Uttam Kumaran: I I think, just looking at this. It looks like we’ve also.
100 00:14:35.630 ⇒ 00:14:50.699 Uttam Kumaran: you know, which is great. We’ve done some some work around refactoring, but I just want to really be careful that it’s just gonna become way too much to handle. There’s 23 models, and there’s a lot I want to have a really good plan before we do any significant refactoring.
101 00:14:51.126 ⇒ 00:14:57.049 Uttam Kumaran: But maybe let’s chat offline. I didn’t know we were actually the original ask was just to move these over.
102 00:14:58.210 ⇒ 00:14:59.090 Luke Daque: Yeah, sure.
103 00:14:59.090 ⇒ 00:15:03.598 Uttam Kumaran: Okay, so let’s chat offline. I I just wanna make sure that that happens.
104 00:15:04.470 ⇒ 00:15:12.030 Uttam Kumaran: you know, before we make decisions on all these staging tables. I don’t wanna. I don’t wanna follow line for line what already exists.
105 00:15:12.290 ⇒ 00:15:13.470 Uttam Kumaran: because
106 00:15:13.580 ⇒ 00:15:21.669 Uttam Kumaran: again, as you can see here, we already have, like 50 or 100 tables. This is way too much for me to kind of comprehend and and understand right now. So
107 00:15:22.111 ⇒ 00:15:27.259 Uttam Kumaran: yeah, let’s chat offline about this, but that’s sort of the entirety of the Dbt work right now.
108 00:15:29.530 ⇒ 00:15:35.630 Bo Yoon: So just I’m not really familiar with all of these, the Dvt are we just
109 00:15:36.110 ⇒ 00:15:43.480 Bo Yoon: updating the big queries tables? The data tables, said, what we’re doing here.
110 00:15:43.780 ⇒ 00:15:47.610 Uttam Kumaran: Yeah. So one is, we’re just moving the logic that is
111 00:15:48.270 ⇒ 00:15:50.400 Uttam Kumaran: as part of those saved queries
112 00:15:50.520 ⇒ 00:16:01.464 Uttam Kumaran: into where it’s like, basically really easily parsable and understandable. If I can show github, you can actually see these queries.
113 00:16:02.110 ⇒ 00:16:05.880 Uttam Kumaran: I don’t think we’ve actually pushed everything yet. Actually, let me just go to
114 00:16:09.580 ⇒ 00:16:12.390 Uttam Kumaran: Are these in a branch yet? Ryan?
115 00:16:12.790 ⇒ 00:16:16.999 Luke Daque: It’s in that branch the Dbt in it, cause I don’t have access to
116 00:16:17.650 ⇒ 00:16:23.770 Luke Daque: like edit, I mean, push anything or create a branch in Github yet, so I just use the cloud
117 00:16:24.702 ⇒ 00:16:30.839 Luke Daque: Dbt. Cloud and it, and we can’t change any branch there as well. So I just use. Whatever branch was available.
118 00:16:31.340 ⇒ 00:16:37.510 Uttam Kumaran: Alright? So yeah, basically, we’ll be able to see all the logic really clearly in one area.
119 00:16:38.118 ⇒ 00:16:47.259 Uttam Kumaran: and basically, everything is version controlled now, and if we wanted to make like a logic change, we can go make it here, and it gets picked up and executed instead of living in
120 00:16:47.500 ⇒ 00:16:51.479 Uttam Kumaran: one or several different you know safe queries.
121 00:16:51.800 ⇒ 00:16:52.629 Robert Tseng: Yeah, so.
122 00:16:52.630 ⇒ 00:16:53.250 Bo Yoon: Fuck.
123 00:16:53.610 ⇒ 00:16:54.139 Robert Tseng: Oh, yeah. Yeah.
124 00:16:54.140 ⇒ 00:16:54.840 Bo Yoon: Go ahead!
125 00:16:55.340 ⇒ 00:17:18.060 Robert Tseng: You know, like, you know, we, we get a bunch of questions daily where it’s really just like we need to update the calculation for a metric or add something in. But then Rob is the only one that touches the schedule queries right now. And like this is just gonna make it easier for us to collaborate and and give the stakeholders more visibility into like. How.
126 00:17:18.069 ⇒ 00:17:18.699 Bo Yoon: Oh, okay.
127 00:17:18.700 ⇒ 00:17:24.923 Robert Tseng: Actually modeling out these different metrics. Because right now, like Rob’s process is the bottleneck and
128 00:17:25.430 ⇒ 00:17:27.839 Robert Tseng: yeah, for us to be able to
129 00:17:28.240 ⇒ 00:17:36.429 Robert Tseng: collaborate with him easier, like, we kind of need a version control environment. So Dbt is, gonna help us do that faster.
130 00:17:37.530 ⇒ 00:17:39.550 Bo Yoon: Okay, got it, got it.
131 00:17:41.690 ⇒ 00:17:47.959 Uttam Kumaran: And you know, for your work. Also, you hopefully, you’ll you’ll have some clear tables where you can go to find the source of truth, for
132 00:17:48.120 ⇒ 00:17:52.530 Uttam Kumaran: users spend all their data. Instead of like
133 00:17:52.950 ⇒ 00:17:57.420 Uttam Kumaran: having having to write these large sort of queries every time you need to find something.
134 00:17:57.420 ⇒ 00:17:58.809 Bo Yoon: Am I still connected.
135 00:17:59.190 ⇒ 00:18:00.290 Robert Tseng: Yeah, you’re still here.
136 00:18:01.140 ⇒ 00:18:01.760 Uttam Kumaran: Yeah.
137 00:18:02.070 ⇒ 00:18:03.472 Bo Yoon: Oh, yeah, sorry.
138 00:18:04.600 ⇒ 00:18:07.879 Bo Yoon: My connection is not stable. Were you saying something.
139 00:18:07.880 ⇒ 00:18:23.460 Uttam Kumaran: No, I was. Yeah. I was. Gonna say that you you’ll actually be able to pull from pretty Consolidated like user tables, spend tables instead of like writing these large queries. Every single time you need one piece, there will be pretty defined areas to get those metrics that are signed off by the business.
140 00:18:23.890 ⇒ 00:18:25.589 Bo Yoon: Oh, okay, I see. See?
141 00:18:27.130 ⇒ 00:18:33.069 Robert Tseng: Yeah, cause. Well, I don’t know about you. But every time I’m writing a query right now, I basically take a snippet from like.
142 00:18:33.330 ⇒ 00:18:59.230 Robert Tseng: or if there is already materialized view, then I have to create, like temp tables or cts on top of that to get to the transformation I want. And it’s like, yeah, the idea is that we would be able to just pull from a a single table for for particular, for a set of use cases. That doesn’t require us to like inject our own custom code every time. In order to get it to where we want it to be.
143 00:19:00.870 ⇒ 00:19:01.840 Bo Yoon: Okay.
144 00:19:03.540 ⇒ 00:19:08.380 Bo Yoon: So the the way that we pull data, I mean, for example, I just
145 00:19:09.180 ⇒ 00:19:15.109 Bo Yoon: what I’ve been doing is just downloading the whole data from the orders table? Or is detail table?
146 00:19:15.530 ⇒ 00:19:19.119 Bo Yoon: Is this gonna is this, gonna make that easier, like.
147 00:19:19.120 ⇒ 00:19:27.920 Uttam Kumaran: Yeah. So you’ll be able to query the orders table directly, and sort of not worry about it like changing, and also have pretty well defined
148 00:19:28.220 ⇒ 00:19:37.809 Uttam Kumaran: like columns there. And anytime you’re like, Hey, I want this new column added. There’s a process that that can do that without breaking. You know everything downstream.
149 00:19:38.170 ⇒ 00:19:41.600 Bo Yoon: So that’s sort of this. And again, we’re just talking about orders.
150 00:19:41.700 ⇒ 00:19:44.829 Uttam Kumaran: Think about this, for you know all core.
151 00:19:44.950 ⇒ 00:19:47.200 Uttam Kumaran: You know business objects.
152 00:19:50.250 ⇒ 00:19:59.219 Bo Yoon: so so when I, whenever I want to pull the data, do i make the sequel query here in dvt or in in bigquery.
153 00:20:00.220 ⇒ 00:20:26.109 Uttam Kumaran: So it it sort of depends the kind of analysis you’re doing. If you’re just pulling data for like a basic analysis, I would do it here in bigquery. And Dbt is where I would work in Dbt, if you’re trying to create a new model or create a new like table that you want to be part of the data model. And that’s like where we would work in Dbt to sort of get that code committed. But all of your preliminary analysis. Just do all that. And
154 00:20:26.260 ⇒ 00:20:34.329 Uttam Kumaran: in bigquery, if you find yourself like running the same query a ton of times, or you’re like, hey? I would love to have this sort of analysis materialized on a schedule.
155 00:20:34.690 ⇒ 00:20:36.610 Uttam Kumaran: That’s where we would do it in Dbt.
156 00:20:36.610 ⇒ 00:20:38.540 Bo Yoon: Okay, okay, got it.
157 00:20:39.380 ⇒ 00:20:40.557 Bo Yoon: Yeah, sounds good.
158 00:20:52.440 ⇒ 00:21:05.599 Nicolas Sucari: Thanks for sharing. Well, maybe, can you share what you’ve been doing on the cohort analysis and so that we can see the visualization tool that you’re using. Maybe you have any question there, and we can help you
159 00:21:05.750 ⇒ 00:21:06.580 Nicolas Sucari: answered that one.
160 00:21:06.580 ⇒ 00:21:12.389 Bo Yoon: Oh, yeah, I’m I’m I’m actually still working on cleaning the the code. I was. I was.
161 00:21:12.630 ⇒ 00:21:21.769 Bo Yoon: Yeah, I actually totally forgot about this meeting before. I’ll get that call ready by our meeting. At one we have a meeting of one right.
162 00:21:22.260 ⇒ 00:21:23.020 Nicolas Sucari: Yeah, yeah.
163 00:21:23.250 ⇒ 00:21:25.813 Bo Yoon: Yeah, I’ll get that done by one
164 00:21:26.580 ⇒ 00:21:30.870 Bo Yoon: I think I showed everything. Do do you need to see the the dashboard as well.
165 00:21:31.810 ⇒ 00:21:41.329 Nicolas Sucari: And no, I was just wondering if you wanted to share that with autumn and Robert, so that they can see what the visualization tool you’re using just in case.
166 00:21:41.330 ⇒ 00:21:42.290 Bo Yoon: Yeah, we need to.
167 00:21:42.540 ⇒ 00:21:44.880 Bo Yoon: So let me let me show you my.
168 00:21:45.040 ⇒ 00:21:49.336 Bo Yoon: the dashboard that I built here. I think.
169 00:22:10.170 ⇒ 00:22:16.190 Bo Yoon: okay, so this is a custom dashboard that I built with python.
170 00:22:16.950 ⇒ 00:22:23.580 Bo Yoon: So I basically downloaded the the whole data table from the orders, table orders, detail table
171 00:22:24.090 ⇒ 00:22:27.400 Bo Yoon: and did all the the pre-processing
172 00:22:29.380 ⇒ 00:22:33.700 Bo Yoon: use some custom functions to to make all of this.
173 00:22:35.040 ⇒ 00:22:41.539 Bo Yoon: But yeah, as I as I told Robert and Nicholas.
174 00:22:42.740 ⇒ 00:22:48.830 Bo Yoon: I don’t. I don’t think this is something that we can do it with the the looker studio or mix panel.
175 00:22:49.490 ⇒ 00:22:51.080 Bo Yoon: So I was.
176 00:22:51.320 ⇒ 00:22:59.779 Bo Yoon: And I’m not really good with deploying stuff like this. I can. I can make this work in my local computer. But
177 00:23:00.000 ⇒ 00:23:01.880 Bo Yoon: deploying it was.
178 00:23:02.180 ⇒ 00:23:08.079 Bo Yoon: It’s a different thing that I I was struggling to get.
179 00:23:08.080 ⇒ 00:23:11.930 Uttam Kumaran: Are you using python like a what python library are you using for, viz.
180 00:23:12.830 ⇒ 00:23:15.149 Bo Yoon: For this dash.
181 00:23:16.700 ⇒ 00:23:22.800 Uttam Kumaran: Okay, okay? And then, are you? This isn’t in a note like, Are you opposed to running this in a notebook.
182 00:23:24.570 ⇒ 00:23:32.810 Bo Yoon: In the notebook. I mean initially, I did it in a notebook. But to to deploy this here to just use a python file.
183 00:23:33.080 ⇒ 00:23:33.890 Uttam Kumaran: Okay. Okay.
184 00:23:33.890 ⇒ 00:23:38.560 Bo Yoon: Yeah, so so everything is python for me as of now.
185 00:23:38.560 ⇒ 00:23:39.540 Uttam Kumaran: Okay. Okay.
186 00:23:39.540 ⇒ 00:23:40.160 Bo Yoon: Yeah.
187 00:23:40.400 ⇒ 00:23:48.260 Bo Yoon: So I was, I I needed to. My code right now is kind of messy. So I need to. I need some time to clean it.
188 00:23:48.450 ⇒ 00:23:52.460 Bo Yoon: and I was going to hand that over to Nicholas and Luke
189 00:23:52.610 ⇒ 00:23:54.030 Bo Yoon: so they could take a look.
190 00:23:56.400 ⇒ 00:24:01.170 Bo Yoon: Yeah, and I guess with Dvt. We can make this
191 00:24:01.810 ⇒ 00:24:04.980 Bo Yoon: easier to to be deployed on the cloud.
192 00:24:06.400 ⇒ 00:24:13.730 Bo Yoon: Yeah, I I will. I mean, I think we probably need to. If this is an app like a python app. We can probably help you deploy it on Gcp.
193 00:24:14.830 ⇒ 00:24:17.330 Bo Yoon: Gcp. What was is that?
194 00:24:17.330 ⇒ 00:24:19.730 Uttam Kumaran: Like, on, on, Google, Cloud.
195 00:24:19.730 ⇒ 00:24:23.440 Bo Yoon: On Google Cloud. Yeah, this is this is actually deployed on Google Cloud. Here.
196 00:24:23.440 ⇒ 00:24:25.880 Uttam Kumaran: Oh, okay, I mean, then I I guess
197 00:24:26.180 ⇒ 00:24:27.790 Uttam Kumaran: I don’t know like, what’s the problem.
198 00:24:28.140 ⇒ 00:24:35.189 Bo Yoon: The problem will be connecting the data directly to this app, like from the sequel database.
199 00:24:35.850 ⇒ 00:25:03.110 Nicolas Sucari: Right. Now the process is that downloads the Csv file directly from Bigquery. Then do some local transformation and then uploads a new Csv with that transformation into into this and deploys. So what we were trying to do is try to understand what are those transformations that he has in his python script create some modeling in Dbt, so that that it’s accessible
200 00:25:03.180 ⇒ 00:25:13.120 Nicolas Sucari: and can like update the data without doing it manually, that we can deploy with better updates this this report.
201 00:25:13.350 ⇒ 00:25:19.019 Bo Yoon: Yeah, so it could basically automatically update every time, like every every day or so.
202 00:25:19.640 ⇒ 00:25:26.849 Uttam Kumaran: Yeah, I mean, that’s definitely possible, like, we could just get all this into a ticket like I can share how to make that happen.
203 00:25:27.740 ⇒ 00:25:29.300 Bo Yoon: Yeah, yeah, sure.
204 00:25:29.300 ⇒ 00:25:40.139 Uttam Kumaran: Okay, yeah. As long as as long as you’re already deployed on. Gcp, I can show you. We’ll just connect to bigquery via like python Api, and I’ll show you how to grab data. But yeah, all this
205 00:25:40.560 ⇒ 00:25:43.600 Uttam Kumaran: document somewhere. Then, yeah, I can suggest how to make that happen.
206 00:25:43.890 ⇒ 00:25:48.190 Uttam Kumaran: And if we can get access also to the file, like, wherever
207 00:25:48.320 ⇒ 00:25:51.619 Uttam Kumaran: in Gcp, you’re deploying this resource, then I can help.
208 00:25:52.260 ⇒ 00:25:59.470 Bo Yoon: Yeah, I actually deployed this to my personal account because I didn’t want to mess with the with a corporate account.
209 00:25:59.670 ⇒ 00:26:09.500 Bo Yoon: But I have all the data. I have all the code here. Just need some time to clean it. I’ll hand that over to Nicholas and Luke in about a few hours.
210 00:26:09.500 ⇒ 00:26:15.930 Uttam Kumaran: Okay, cool. Yeah. I think at minimum. If you could just shove every as much as possible into like a collab on the company.
211 00:26:16.050 ⇒ 00:26:19.900 Uttam Kumaran: then we can work on, probably moving that colab into something that’s like.
212 00:26:20.230 ⇒ 00:26:22.670 Uttam Kumaran: you know, like a basic app. Probably there.
213 00:26:23.320 ⇒ 00:26:25.530 Bo Yoon: A call app. Oh, what do you mean by call it.
214 00:26:25.530 ⇒ 00:26:28.200 Uttam Kumaran: Like a notebook like a Google colab notebook.
215 00:26:29.020 ⇒ 00:26:31.130 Bo Yoon: Oh, you want me to. Okay, yeah, sure.
216 00:26:31.130 ⇒ 00:26:37.240 Uttam Kumaran: Well, like, yeah, I guess I you just mentioned that it’s like this is on your personal. So I wanna get it into there
217 00:26:37.940 ⇒ 00:26:44.160 Uttam Kumaran: ecosystem, like, I think, Google Collabs, probably the best place to just put it a less.
218 00:26:44.160 ⇒ 00:26:44.660 Bo Yoon: Hmm.
219 00:26:44.660 ⇒ 00:26:47.339 Uttam Kumaran: You wanna send me a repo or something.
220 00:26:48.460 ⇒ 00:26:52.200 Bo Yoon: Okay, okay, got it? Yeah, let me let me let me figure that out.
221 00:26:57.630 ⇒ 00:27:05.020 Bo Yoon: Okay, so are you are.
222 00:27:05.930 ⇒ 00:27:07.140 Bo Yoon: It’s Robert.
223 00:27:07.290 ⇒ 00:27:13.880 Bo Yoon: Are are you all working for Brainforge, or said, I’m not sure how this works here.
224 00:27:13.880 ⇒ 00:27:18.790 Uttam Kumaran: I would just like I would just communicate directly with Nico, and he’ll he’ll collect everything.
225 00:27:19.590 ⇒ 00:27:20.950 Bo Yoon: With Nico. Okay?
226 00:27:20.950 ⇒ 00:27:21.490 Nicolas Sucari: Kim.
227 00:27:21.760 ⇒ 00:27:22.480 Bo Yoon: Got it.
228 00:27:28.190 ⇒ 00:27:32.041 Bo Yoon: so I’ll get help deploying this on the cloud.
229 00:27:32.590 ⇒ 00:27:37.280 Bo Yoon: And and this is just I’ll need to add a lot of more features to this app.
230 00:27:37.870 ⇒ 00:27:42.719 Bo Yoon: So this is, gonna get a lot more complicated.
231 00:27:43.380 ⇒ 00:27:44.860 Bo Yoon: And what we see here.
232 00:27:46.010 ⇒ 00:27:51.190 Uttam Kumaran: Yup, yeah, we’ll we’ll I can, you know. We’ll assist on how to how to make all that happen again.
233 00:27:51.190 ⇒ 00:27:51.680 Bo Yoon: Hmm.
234 00:27:51.680 ⇒ 00:28:00.149 Uttam Kumaran: As much as we can just download and get into something structured where you know, and I can assist the our team will assist to help you make that happen. So.
235 00:28:00.680 ⇒ 00:28:02.421 Bo Yoon: Yeah, that would be great.
236 00:28:02.770 ⇒ 00:28:12.109 Nicolas Sucari: 1st step is to go through the the script in the next meeting. Oh, once we have that, we can start hoping on how to deploy this and to improve it.
237 00:28:12.780 ⇒ 00:28:14.430 Bo Yoon: Okay, yeah, that would be great.
238 00:28:15.290 ⇒ 00:28:17.119 Bo Yoon: Thank you. Guys.
239 00:28:17.700 ⇒ 00:28:20.800 Nicolas Sucari: Thank you. Anything else. Guys.
240 00:28:26.380 ⇒ 00:28:31.189 Robert Tseng: Let’s let’s chat about like what? What needs to change about this check in for future ones.
241 00:28:33.390 ⇒ 00:28:42.550 Robert Tseng: So like we who who else needs to be on on this call like we, we should kind of figure out like what? What cadence we’re meeting at like this is supposed to be like
242 00:28:42.680 ⇒ 00:28:51.199 Robert Tseng: the one check in that the brain force folks have with the with the Eden folks, so that if, whatever you, whatever we need, we we should just talk about what what that looks like.
243 00:28:51.580 ⇒ 00:28:53.959 Robert Tseng: and I’ll figure out how to get the right people in there.
244 00:28:53.960 ⇒ 00:29:02.950 Uttam Kumaran: Yeah, I think for me like it wouldn’t. I just wanted to talk to Rob, just to say Hi and but I don’t really need. I don’t really have any needs at the moment.
245 00:29:03.296 ⇒ 00:29:07.789 Uttam Kumaran: And then, yeah, I just once you get the requirements for this Nico, I can take a look, but
246 00:29:07.940 ⇒ 00:29:09.799 Uttam Kumaran: probably don’t need to chat before then.
247 00:29:11.090 ⇒ 00:29:15.960 Bo Yoon: Yeah, sure. Should we make this like a weekly meeting including Rob?
248 00:29:17.410 ⇒ 00:29:19.649 Bo Yoon: So so you can join if you can.
249 00:29:19.650 ⇒ 00:29:22.140 Uttam Kumaran: Yeah, is there anyone else? Beau on the data side?
250 00:29:22.310 ⇒ 00:29:24.460 Bo Yoon: The data side there is.
251 00:29:25.520 ⇒ 00:29:28.360 Bo Yoon: there’s only Zack, but I’m I’m not sure.
252 00:29:28.460 ⇒ 00:29:34.550 Bo Yoon: I think he only works with the luquer studio. I’m I’m not sure, Robert. Do you know.
253 00:29:35.433 ⇒ 00:29:41.636 Robert Tseng: Yeah. Well, Zack is. I mean, he’s pretty much like not doing data stuff anymore.
254 00:29:42.050 ⇒ 00:29:42.870 Bo Yoon: Oh, okay.
255 00:29:43.250 ⇒ 00:30:01.040 Robert Tseng: Yeah, he like manages Google tag manager, and like, he’ll add new pixels. And whenever marketing is like jumping at the next new tool, he’ll like set up connection. But that’s it. He just says that. I I think I also want to bring in Sebastian, who, I guess, is the new CTO that they just hired
256 00:30:01.753 ⇒ 00:30:16.009 Robert Tseng: onto the call as well. Just because I think he’ll want to know about the tool stack that we’re using, and he’ll have some questions about how we’re like managing, how we’re managing the data and and what kind of controls we’re putting in place.
257 00:30:16.481 ⇒ 00:30:35.659 Robert Tseng: But yeah, I’ll make sure that Rob’s on on the call moving forward. Because, yeah, I mean, like, I just, I just wanna get a good snapshot where we talk about everything, data, priority wise that’s in flight, so that we can like be bit more coordinated about
258 00:30:35.800 ⇒ 00:30:49.139 Robert Tseng: whose work is supporting, who’s work downstream and and whatnot, because I think everyone’s used to operating pretty independently. But there are some dependencies that I think would be faster to clear up if we have this type of check in.
259 00:30:50.850 ⇒ 00:30:57.339 Bo Yoon: Yeah, yeah, that would be great. I I still have a lot of things that I don’t know about this company. So
260 00:30:58.400 ⇒ 00:31:00.279 Bo Yoon: yeah, sounds great.
261 00:31:02.760 ⇒ 00:31:04.060 Robert Tseng: Okay. Cool.
262 00:31:05.980 ⇒ 00:31:12.820 Bo Yoon: Sure. So should we make this a weekly meeting at this time? Or if that’s okay with you.
263 00:31:13.150 ⇒ 00:31:15.440 Robert Tseng: Yeah, I mean that that works for me.
264 00:31:17.900 ⇒ 00:31:27.059 Nicolas Sucari: Yeah, I think that’s fine. Let me check with Rob if he’s gonna be able to join at this time, and we can schedule it for every week. If not, we’re gonna find a different time. Okay.
265 00:31:28.140 ⇒ 00:31:28.480 Robert Tseng: Okay.
266 00:31:29.330 ⇒ 00:31:30.680 Bo Yoon: Yeah. Sounds. Great.
267 00:31:31.130 ⇒ 00:31:31.859 Robert Tseng: All right.
268 00:31:32.050 ⇒ 00:31:32.930 Robert Tseng: Hi, everyone!
269 00:31:33.210 ⇒ 00:31:33.850 Bo Yoon: Alright.
270 00:31:33.850 ⇒ 00:31:35.110 Bo Yoon: Yeah, thank you.
271 00:31:35.110 ⇒ 00:31:36.980 Bo Yoon: Have a nice day. Thank you.