Meeting Title: Eden __ Brainforge - Data Modeling Requests Date: 2025-01-08 Meeting participants: Luke Daque, Nicolas Sucari, Bo Yoon
WEBVTT
1 00:02:49.630 ⇒ 00:02:50.620 Nicolas Sucari: Hey! Luke!
2 00:02:52.380 ⇒ 00:02:54.560 Luke Daque: Hi, Nico! How’s it going.
3 00:02:56.390 ⇒ 00:02:58.029 Nicolas Sucari: Good! How are you?
4 00:02:59.190 ⇒ 00:03:00.360 Luke Daque: Yeah. Doing well.
5 00:03:03.570 ⇒ 00:03:06.100 Luke Daque: you happy to be back home.
6 00:03:06.740 ⇒ 00:03:10.280 Nicolas Sucari: Yeah, in a long trip yesterday.
7 00:03:12.660 ⇒ 00:03:15.270 Luke Daque: Nice! How long was it like? How.
8 00:03:16.720 ⇒ 00:03:25.100 Nicolas Sucari: To us like 5 h drive, and then 1 h on a boat with the car.
9 00:03:26.820 ⇒ 00:03:27.420 Nicolas Sucari: Hi, Bill!
10 00:03:27.420 ⇒ 00:03:28.310 Luke Daque: Pretty long.
11 00:03:28.310 ⇒ 00:03:29.180 Bo Yoon: Hello!
12 00:03:29.510 ⇒ 00:03:29.909 Luke Daque: Nice to meet you.
13 00:03:29.910 ⇒ 00:03:30.550 Bo Yoon: Right.
14 00:03:31.260 ⇒ 00:03:33.000 Nicolas Sucari: Nice to meet you, too. How are you.
15 00:03:33.620 ⇒ 00:03:35.270 Bo Yoon: Good! Good! How’s it going.
16 00:03:37.700 ⇒ 00:03:38.280 Luke Daque: So far.
17 00:03:38.280 ⇒ 00:03:39.999 Nicolas Sucari: So good for me today.
18 00:03:40.770 ⇒ 00:03:41.340 Bo Yoon: Yeah.
19 00:03:43.490 ⇒ 00:03:44.170 Bo Yoon: Alright.
20 00:03:44.170 ⇒ 00:03:44.870 Nicolas Sucari: Okay. Great.
21 00:03:44.870 ⇒ 00:03:45.210 Bo Yoon: Nice.
22 00:03:45.210 ⇒ 00:03:46.019 Nicolas Sucari: Nice to meet you.
23 00:03:47.000 ⇒ 00:03:52.479 Bo Yoon: Yeah, very nice to meet you, is, is there anyone else joining us? Or it’s just gonna be us.
24 00:03:52.480 ⇒ 00:03:59.300 Nicolas Sucari: I I think it’s gonna be us. I invited you, Tom, but I’m not sure if he’s gonna join
25 00:04:00.200 ⇒ 00:04:00.610 Bo Yoon: Okay.
26 00:04:00.610 ⇒ 00:04:03.140 Nicolas Sucari: Yeah, we can get right at it.
27 00:04:03.768 ⇒ 00:04:06.339 Nicolas Sucari: So yeah, I was. I was
28 00:04:06.540 ⇒ 00:04:14.089 Nicolas Sucari: trying to understand a little bit on the requirement from Robert. I know you’ve been working on cohort analysis right.
29 00:04:15.120 ⇒ 00:04:15.440 Bo Yoon: Let me!
30 00:04:15.440 ⇒ 00:04:36.239 Nicolas Sucari: The idea my idea today with look here, is to understand a little bit on how you were doing all of those analysis. Where are you getting the data from, and try to explain us, or or do a walkthrough on what you have, so that we can help with the data modeling for you, so that that analysis came comes easily. Maybe.
31 00:04:36.440 ⇒ 00:04:41.620 Bo Yoon: Yeah, yeah, sure. But before we we get started on that, can you?
32 00:04:41.750 ⇒ 00:04:46.970 Bo Yoon: Could you explain a little bit about what brain porch is. Is this like a
33 00:04:47.275 ⇒ 00:04:49.410 Bo Yoon: don’t understand? What’s what’s going on here?
34 00:04:49.630 ⇒ 00:04:50.030 Bo Yoon: Yeah.
35 00:04:50.710 ⇒ 00:05:15.989 Nicolas Sucari: I thought I thought you were aware of that. Sure. We are a data and AI agency we’re working with. Well, Eden, now, right now doing some data and AI stuff. Well, actually, with you guys, it’s just data we are trying to work closely with you and Rob from from Eden’s team. We’re gonna be working on
36 00:05:16.436 ⇒ 00:05:22.689 Nicolas Sucari: rethinking the strategy regarding all of the data from the from the product. Okay.
37 00:05:25.700 ⇒ 00:05:54.959 Nicolas Sucari: We kind of manage all of the data pipeline and flows for different clients. And what we are aiming to do here is to work on stuff regarding product analytics first, st and then I’m not sure if we’re gonna keep working after that. But the the main goal is to help on the data, modeling, sourcing, creating the the correct either workflows or that kind of stuff in order to to have better insights for product analytics.
38 00:05:56.990 ⇒ 00:06:00.438 Bo Yoon: Hmm, okay, so so you basically help us
39 00:06:01.080 ⇒ 00:06:04.289 Bo Yoon: getting the data modeling that I’m doing
40 00:06:04.640 ⇒ 00:06:09.760 Bo Yoon: deploying that on on the cloud. Is that how I should interpret it!
41 00:06:10.910 ⇒ 00:06:27.910 Nicolas Sucari: Yeah, yeah, I think that’s correct. I mean, we help on all of the that engineering stuff. And then also on the analysis, too. Here Luke, is our data engineer, and he’ll be working on everything regarding modeling. So yeah.
42 00:06:28.530 ⇒ 00:06:36.650 Nicolas Sucari: the idea is to help you guys to have better better stuff, so that then we can create better reporting or analysis.
43 00:06:37.810 ⇒ 00:06:45.450 Bo Yoon: Okay, so are, are you guys, are you good with building like custom that are reporting
44 00:06:45.760 ⇒ 00:06:46.970 Bo Yoon: kind of apps.
45 00:06:49.860 ⇒ 00:06:58.510 Nicolas Sucari: We are. I think we’re aiming to move. Move stuff through snowflake and dbt, so yeah.
46 00:06:58.510 ⇒ 00:06:59.289 Bo Yoon: Oh no!
47 00:06:59.290 ⇒ 00:07:02.060 Nicolas Sucari: Kind of yeah, I mean.
48 00:07:02.060 ⇒ 00:07:03.260 Bo Yoon: That’s what we’re gonna be.
49 00:07:03.920 ⇒ 00:07:08.450 Nicolas Sucari: I’m not sure if Snowflake or bigquery I mean, I know you guys will be using bigquery right.
50 00:07:08.990 ⇒ 00:07:12.735 Bo Yoon: Yeah. Yeah. So so for now, what I’m doing is,
51 00:07:14.756 ⇒ 00:07:24.559 Bo Yoon: I’m pulling data from the from the big query on with a SQL. Query. I download the the data in the Csv format. And then I do the
52 00:07:24.730 ⇒ 00:07:27.660 Bo Yoon: the data analysis in my python notebook.
53 00:07:28.440 ⇒ 00:07:34.180 Bo Yoon: And then that’s how I I’ve been generating reports and and all this stuff.
54 00:07:34.620 ⇒ 00:07:41.480 Bo Yoon: Another thing that I it was let me show you what I
55 00:07:44.010 ⇒ 00:07:49.569 Bo Yoon: let me show you dash not sure where that is.
56 00:08:00.310 ⇒ 00:08:05.610 Nicolas Sucari: Yeah, sorry. I know I mentioned Snowflake, but now we are working with Bigquery and Github to do all of the.
57 00:08:05.610 ⇒ 00:08:06.100 Bo Yoon: Bigquery.
58 00:08:06.100 ⇒ 00:08:09.750 Nicolas Sucari: Engineering and modeling. Yeah, bigquery. Exactly. That’s what you’re using.
59 00:08:09.750 ⇒ 00:08:16.480 Bo Yoon: Right? Okay, yeah. So, boy, this is taking a long time. Okay.
60 00:08:17.090 ⇒ 00:08:18.920 Bo Yoon: let me share that with you.
61 00:08:23.740 ⇒ 00:08:28.929 Bo Yoon: Okay, yeah. So so this is the custom dashboard that I built.
62 00:08:31.070 ⇒ 00:08:37.340 Bo Yoon: This is using a few functions that I created myself.
63 00:08:39.309 ⇒ 00:08:47.669 Bo Yoon: But I I don’t think we can. So so at Eden. I think we’re basically using mixed panel and looker studio
64 00:08:48.160 ⇒ 00:09:00.709 Bo Yoon: to generate reports. But I I think there are limitations on that, because I’m I’m using all these python libraries like, for example, like linear regression profit. And all this
65 00:09:00.820 ⇒ 00:09:09.720 Bo Yoon: like machine learning libraries that I I don’t think is available on looker or in mixed panel.
66 00:09:10.490 ⇒ 00:09:12.230 Bo Yoon: So I’m
67 00:09:12.880 ⇒ 00:09:19.429 Bo Yoon: the the data analysis that I’m doing. I don’t think is possible on either of those 2 platforms. So
68 00:09:19.610 ⇒ 00:09:22.400 Bo Yoon: I’ve been building that here.
69 00:09:23.110 ⇒ 00:09:28.330 Bo Yoon: But i i i don’t really have trouble
70 00:09:29.657 ⇒ 00:09:41.360 Bo Yoon: showing it locally in my computer. But when it comes to deploying it in in the cloud here, for example, this one. It took me a while to do this because I’ve been getting so so many errors
71 00:09:42.000 ⇒ 00:09:44.649 Bo Yoon: trying to uploading it to Google Cloud.
72 00:09:46.310 ⇒ 00:09:46.930 Nicolas Sucari: Okay.
73 00:09:47.220 ⇒ 00:09:53.539 Bo Yoon: Yeah. Yeah. And and that’s I, I think that’s where if you could help me with this, that would be great
74 00:09:55.610 ⇒ 00:10:01.650 Bo Yoon: for for this code here. I also needed to. It’ll be great if I could just
75 00:10:01.800 ⇒ 00:10:04.309 Bo Yoon: get the data straight from bigquery
76 00:10:05.250 ⇒ 00:10:09.730 Bo Yoon: and then do the analysis. But that’s
77 00:10:11.050 ⇒ 00:10:19.439 Bo Yoon: not sure how that’s gonna be efficient. Because what you see here is the data that I
78 00:10:20.330 ⇒ 00:10:25.218 Bo Yoon: I pre-process the data first, st and then I download
79 00:10:26.050 ⇒ 00:10:32.439 Bo Yoon: I have a saved in a in a very small Csv file. And what you’re what you’re seeing here is just
80 00:10:33.030 ⇒ 00:10:38.110 Bo Yoon: it’s it’s all coming from a very small size of Csv file.
81 00:10:38.360 ⇒ 00:10:39.170 Bo Yoon: It’s not used.
82 00:10:39.170 ⇒ 00:10:39.830 Nicolas Sucari: Okay.
83 00:10:41.196 ⇒ 00:10:45.650 Bo Yoon: But but if I, if we
84 00:10:46.380 ⇒ 00:10:55.470 Bo Yoon: query the the whole data from from the from the bigquery database is going to be like a large file that we’ll need to
85 00:10:58.300 ⇒ 00:11:00.799 Bo Yoon: preprocess every time. I’m not sure.
86 00:11:01.940 ⇒ 00:11:05.479 Bo Yoon: Are, are you guys, do you guys have any idea for these.
87 00:11:08.180 ⇒ 00:11:18.959 Luke Daque: Yeah, basically, we usually do the transformation like any kind of data transformation or data processing using. Dbt.
88 00:11:19.862 ⇒ 00:11:21.100 Luke Daque: we we yeah.
89 00:11:21.640 ⇒ 00:11:22.180 Bo Yoon: Dvd.
90 00:11:22.180 ⇒ 00:11:24.038 Luke Daque: Yeah, dbt, which is like,
91 00:11:25.020 ⇒ 00:11:35.500 Luke Daque: data build tool, it’s an open source platform. And we just connect it to the bigquery project, basically that already has all the sources data sources.
92 00:11:35.690 ⇒ 00:11:43.199 Luke Daque: And so, yeah, we just need to know, like, how you are like doing the pre-processing
93 00:11:45.710 ⇒ 00:11:46.740 Luke Daque: because it’s I.
94 00:11:46.740 ⇒ 00:11:47.190 Nicolas Sucari: Are you?
95 00:11:47.190 ⇒ 00:11:51.260 Luke Daque: Mentioned you were using, like python, notable, notable.
96 00:11:51.260 ⇒ 00:11:51.790 Nicolas Sucari: Yeah.
97 00:11:51.790 ⇒ 00:11:53.499 Luke Daque: To do the pre-processing right.
98 00:11:54.220 ⇒ 00:11:54.890 Bo Yoon: You can get.
99 00:11:54.890 ⇒ 00:11:55.760 Luke Daque: Like your.
100 00:11:57.320 ⇒ 00:11:58.352 Luke Daque: Yeah. Yeah. Go ahead.
101 00:11:59.490 ⇒ 00:12:03.369 Bo Yoon: So. So yeah, I’m doing everything in python here. If you guys need
102 00:12:04.060 ⇒ 00:12:09.250 Bo Yoon: the script, I can. I can provide you the script that I’m using here for the Pre.
103 00:12:09.250 ⇒ 00:12:12.200 Nicolas Sucari: Do you have that in Github, or did you?
104 00:12:12.200 ⇒ 00:12:13.620 Nicolas Sucari: We don’t need her.
105 00:12:13.840 ⇒ 00:12:17.550 Bo Yoon: No, not yet. I just have a lot saved it locally.
106 00:12:18.710 ⇒ 00:12:23.199 Bo Yoon: but I’ll I’ll have to clean it a little bit. It’s a mess right now, so.
107 00:12:24.300 ⇒ 00:12:27.899 Luke Daque: No worries. Yeah, it’s always like that. If we if we see.
108 00:12:29.700 ⇒ 00:12:31.500 Bo Yoon: Yeah.
109 00:12:31.500 ⇒ 00:12:43.790 Luke Daque: So, yeah, like, if we, you can help us understand, like what the what the logic is or like, what how you’re doing the pre-processing. Maybe we can translate that into sequel
110 00:12:43.990 ⇒ 00:12:50.780 Luke Daque: and then create a different table in bigquery. That’s just basically the.
111 00:12:50.780 ⇒ 00:12:51.430 Bo Yoon: Create, a.
112 00:12:51.430 ⇒ 00:12:52.370 Luke Daque: Looking for.
113 00:12:52.990 ⇒ 00:12:55.399 Bo Yoon: Oh, actually! Never mind!
114 00:12:55.580 ⇒ 00:13:04.670 Luke Daque: So instead of like saving it into a Csv file, we’ll create a just like a a model, a data model or a table in bigquery. Maybe.
115 00:13:04.670 ⇒ 00:13:05.850 Bo Yoon: Another table.
116 00:13:06.160 ⇒ 00:13:08.380 Luke Daque: Yeah, maybe we yeah, yeah, that was data set
117 00:13:08.610 ⇒ 00:13:14.640 Luke Daque: that has all the tables that you need for that’s already processed data. Basically.
118 00:13:15.688 ⇒ 00:13:20.609 Bo Yoon: Yeah, that’ll that would be a great idea just getting a new table.
119 00:13:21.510 ⇒ 00:13:30.000 Luke Daque: Yeah, and depending on how often you need the the data updated, whether it’s like once a day or like every 30 min, or whatever.
120 00:13:30.545 ⇒ 00:13:38.809 Luke Daque: We can automate that as well like. Have dbt, run this the the SQL. Query that will generate that.
121 00:13:38.810 ⇒ 00:13:39.280 Bo Yoon: And.
122 00:13:39.280 ⇒ 00:13:40.710 Luke Daque: Data. And then, yeah.
123 00:13:41.430 ⇒ 00:13:48.380 Bo Yoon: Okay. Another question is that what you’re seeing here is for one product
124 00:13:50.460 ⇒ 00:13:51.190 Nicolas Sucari: Okay.
125 00:13:51.190 ⇒ 00:13:56.330 Bo Yoon: I think we’ll have to do this for every product that we have added in
126 00:13:56.810 ⇒ 00:14:04.329 Bo Yoon: in in that case, do we have to build like multiple tables for each product? Or is that gonna be just.
127 00:14:04.710 ⇒ 00:14:15.160 Luke Daque: May maybe we can consolidate everything into just one table and just have one field that’s product, and then add it as a filter in in this
128 00:14:15.570 ⇒ 00:14:17.360 Luke Daque: report, or something right.
129 00:14:17.970 ⇒ 00:14:21.080 Bo Yoon: Edit as a filter like you can. You can.
130 00:14:21.080 ⇒ 00:14:27.240 Luke Daque: You can just change the product, and then you’ll be able to see the Ltv over time.
131 00:14:29.080 ⇒ 00:14:33.200 Luke Daque: That specific product, unless you need like different.
132 00:14:34.470 ⇒ 00:14:35.010 Nicolas Sucari: Or maybe.
133 00:14:35.010 ⇒ 00:14:35.930 Luke Daque: Data, visualize.
134 00:14:36.250 ⇒ 00:14:36.840 Luke Daque: Just yet.
135 00:14:37.380 ⇒ 00:14:46.770 Nicolas Sucari: Yeah, maybe the model is the same look like if we have, like all of the products in the same model. But then, when we? When we go into the visualization we can filter it down.
136 00:14:46.880 ⇒ 00:14:49.889 Nicolas Sucari: have it differently. Maybe that’s
137 00:14:50.180 ⇒ 00:15:06.740 Nicolas Sucari: possible, too. Bo, can you? I mean, let me see if I kind of understand, like your processes, you, you create a query. For in in bigquery, so that you extract some data that data you exported as Csv, right? Without any transformation. Yet, right?
138 00:15:07.636 ⇒ 00:15:22.550 Nicolas Sucari: It’s raw data from bigquery. Okay, once you have that Csv, you do some python transformation. And you have your python script, and once once you have that you save it again as a Csv file, and you use it as a source for this report. Right? That’s kind of your process.
139 00:15:23.800 ⇒ 00:15:27.763 Bo Yoon: Yeah, for for this one here I tried doing
140 00:15:29.350 ⇒ 00:15:40.399 Bo Yoon: I tried drawing it with the, with the pre-processing data in in the script for this graph as well. But that didn’t work, because I think there were some limitations in the size of the file
141 00:15:40.990 ⇒ 00:15:42.620 Bo Yoon: for for Google Cloud.
142 00:15:43.030 ⇒ 00:15:51.839 Bo Yoon: So yeah, yeah, that’s that’s basically what I did. I, I downloaded the the whole data for the in the order details table.
143 00:15:52.180 ⇒ 00:15:55.020 Bo Yoon: which is the raw data for each customer.
144 00:15:55.550 ⇒ 00:16:01.429 Bo Yoon: and then I pre-process it with python and and everything here is in python as well.
145 00:16:03.490 ⇒ 00:16:21.779 Nicolas Sucari: Okay, I think. Look, what we need is access to that transformations in Python or that python script. And see. Maybe, Bo, you can guide us through what you’re doing, or tell us a little bit what you’re doing on those transformations so that we can have an idea of what is that you’re doing there. And then.
146 00:16:22.120 ⇒ 00:16:27.900 Nicolas Sucari: if we have that script, we can start looking Brian into sorry. Look into how we are.
147 00:16:28.290 ⇒ 00:16:34.389 Nicolas Sucari: how we can like change that or create that into in in DVD, right? So that we can create that model.
148 00:16:36.010 ⇒ 00:16:42.159 Bo Yoon: Yeah, sure. Let’s see. Process there.
149 00:16:47.480 ⇒ 00:16:49.980 Bo Yoon: Okay, let me let me share it with you.
150 00:16:51.730 ⇒ 00:16:57.520 Bo Yoon: I mean, the pre-processing was was very simple. Here, where is this?
151 00:17:10.310 ⇒ 00:17:13.140 Bo Yoon: Okay? So are you guys
152 00:17:13.480 ⇒ 00:17:22.339 Bo Yoon: good with python? Okay? Yeah. It’s very simple. Downloaded the the the whole data here.
153 00:17:22.740 ⇒ 00:17:30.500 Bo Yoon: And Csv file cleaned it a little bit, change the the data type
154 00:17:32.156 ⇒ 00:17:35.840 Bo Yoon: and then what? They, the
155 00:17:37.784 ⇒ 00:17:46.029 Bo Yoon: the product filter. And then, oh, actually, I’m using another another function here.
156 00:17:46.200 ⇒ 00:17:52.749 Bo Yoon: Alright, then, yeah. I’m using another function to to get the the course analysis
157 00:17:53.030 ⇒ 00:18:01.189 Bo Yoon: and the the core table, and then just changing the names here and filtering the
158 00:18:03.650 ⇒ 00:18:05.240 Bo Yoon: Oh, I’m using another.
159 00:18:06.007 ⇒ 00:18:11.180 Bo Yoon: Okay, yeah, I’ll have to share with you all these functions as well.
160 00:18:11.811 ⇒ 00:18:15.880 Bo Yoon: I think I’ll need to clean it a little bit. It’s just a mess here.
161 00:18:16.010 ⇒ 00:18:19.470 Bo Yoon: But but this is basically how I did the pre-processing.
162 00:18:19.690 ⇒ 00:18:21.620 Bo Yoon: So using a few functions
163 00:18:22.819 ⇒ 00:18:29.109 Bo Yoon: to clean the data set and then transforming it to the to a cohort table
164 00:18:30.260 ⇒ 00:18:32.569 Bo Yoon: and then changing the names a little bit
165 00:18:34.193 ⇒ 00:18:39.280 Bo Yoon: calculating the Ltv list is going to be the retention rate here.
166 00:18:40.410 ⇒ 00:18:46.179 Bo Yoon: And yeah, this is it for that.
167 00:18:46.850 ⇒ 00:18:47.618 Luke Daque: Do you have like
168 00:18:47.810 ⇒ 00:18:48.284 Bo Yoon: But
169 00:18:48.760 ⇒ 00:18:54.980 Luke Daque: A sample like how the the final data frame, or like, would would look like like what the
170 00:18:54.980 ⇒ 00:18:56.150 Luke Daque: oh, the r.
171 00:18:57.310 ⇒ 00:18:58.920 Bo Yoon: Like, what? The? Yeah.
172 00:18:59.180 ⇒ 00:19:07.390 Luke Daque: Final table. Is it still in a i i suppose it’s still gonna be saved as Csv, right? Like you mentioned, it’s just a small Csv file.
173 00:19:08.180 ⇒ 00:19:11.119 Bo Yoon: Yeah, it was just a small Csv file. I
174 00:19:11.770 ⇒ 00:19:17.599 Bo Yoon: I worked on this like a few weeks ago. I I need to kind of organize it.
175 00:19:18.650 ⇒ 00:19:20.069 Bo Yoon: Don’t know where that is.
176 00:19:20.750 ⇒ 00:19:25.180 Bo Yoon: 4, 5 months, let me tell.
177 00:19:29.080 ⇒ 00:19:31.889 Luke Daque: So far. Yeah, I think that that should work like
178 00:19:32.070 ⇒ 00:19:35.549 Luke Daque: if you if we get the logic on how
179 00:19:36.380 ⇒ 00:19:41.350 Luke Daque: those cohorts are being calculated, and stuff like that. What the filters you’re doing? Yeah.
180 00:19:41.350 ⇒ 00:19:41.719 Bo Yoon: Didn’t know.
181 00:19:41.720 ⇒ 00:19:49.640 Luke Daque: All the pre-processing stuff. Then maybe we can find a way to do it in Dbt.
182 00:19:50.600 ⇒ 00:19:51.140 Bo Yoon: Indeed.
183 00:19:51.140 ⇒ 00:19:55.619 Bo Yoon: Dvt, okay, is Dvt, just a library or.
184 00:19:56.090 ⇒ 00:19:57.140 Luke Daque: All right.
185 00:19:57.280 ⇒ 00:20:00.376 Luke Daque: No, it’s a, it’s a, it’s an open source.
186 00:20:01.810 ⇒ 00:20:16.170 Luke Daque: data modeling tool, basically. Where you connect it to like a data warehouse like bigquery. And then it does. You can create data models using sequel and ginger, it has ginger functions as well.
187 00:20:17.100 ⇒ 00:20:20.799 Luke Daque: And then you can automate the data transformation where? Like.
188 00:20:20.980 ⇒ 00:20:23.140 Luke Daque: yeah, you can schedule it. Basically.
189 00:20:23.570 ⇒ 00:20:25.550 Bo Yoon: The output would multiply, scheduling.
190 00:20:25.730 ⇒ 00:20:28.750 Luke Daque: Yeah, schedule the the data transformation
191 00:20:30.205 ⇒ 00:20:32.920 Luke Daque: and then output it into a
192 00:20:33.200 ⇒ 00:20:38.899 Luke Daque: another table in bigquery that way. We don’t have like Csv files everywhere.
193 00:20:39.010 ⇒ 00:20:39.740 Luke Daque: It’s still.
194 00:20:40.160 ⇒ 00:20:41.050 Luke Daque: Data, warehouse.
195 00:20:41.050 ⇒ 00:20:43.210 Nicolas Sucari: Yeah, so cheers, how?
196 00:20:43.210 ⇒ 00:20:43.800 Bo Yoon: Okay.
197 00:20:44.160 ⇒ 00:20:52.010 Nicolas Sucari: Yeah, so how? How like, how often do you update that data in order to have like updated values for that report? Or you just.
198 00:20:52.010 ⇒ 00:20:53.069 Bo Yoon: For this one.
199 00:20:54.050 ⇒ 00:21:00.230 Bo Yoon: Yeah. So for this dashboard, Adam and the C level guys there.
200 00:21:01.270 ⇒ 00:21:07.620 Bo Yoon: they’re they’re the only ones using this. I think so. I think once a day should be enough
201 00:21:08.290 ⇒ 00:21:13.249 Bo Yoon: for the data. Update it. It doesn’t really need to be like every 30 min, because that’s
202 00:21:14.750 ⇒ 00:21:16.860 Bo Yoon: you don’t know that that makes sense cost us.
203 00:21:16.860 ⇒ 00:21:17.590 Nicolas Sucari: I’m
204 00:21:17.800 ⇒ 00:21:25.720 Nicolas Sucari: as for now, you were exporting once a day, and like rebuilding the model with the new raw data every day. Or how were you.
205 00:21:25.720 ⇒ 00:21:32.479 Bo Yoon: No, no, not not every day. I I did this like a few few weeks ago. So this is this is
206 00:21:32.810 ⇒ 00:21:35.350 Bo Yoon: this contains the data. Yeah.
207 00:21:35.580 ⇒ 00:21:42.599 Bo Yoon: yeah, from from December. I think December was the last time December mid December was the last time I worked on this.
208 00:21:43.040 ⇒ 00:21:44.970 Bo Yoon: I’ve been working on other things.
209 00:21:45.310 ⇒ 00:21:50.169 Bo Yoon: Is that for for the code, let me.
210 00:21:50.280 ⇒ 00:22:01.210 Bo Yoon: I’m not sure where I saved. This is the the folder. Here is just a mess here. I’ll need some time to organize it. Clean the the code as well. So if you guys
211 00:22:02.960 ⇒ 00:22:06.280 Bo Yoon: do, we wanna have another time, maybe
212 00:22:07.390 ⇒ 00:22:10.930 Bo Yoon: for another meeting. Yeah, is that gonna be okay with you guys.
213 00:22:10.930 ⇒ 00:22:15.302 Luke Daque: Yeah, that’s fine. You can also, just like, probably email,
214 00:22:15.700 ⇒ 00:22:16.629 Bo Yoon: Oh, yeah, yeah, I can.
215 00:22:16.630 ⇒ 00:22:17.659 Luke Daque: I can like to see.
216 00:22:17.660 ⇒ 00:22:19.276 Bo Yoon: To maybe share the code.
217 00:22:19.600 ⇒ 00:22:20.140 Luke Daque: Yeah.
218 00:22:21.060 ⇒ 00:22:22.910 Nicolas Sucari: Or go go through slack. Yeah.
219 00:22:23.430 ⇒ 00:22:23.760 Luke Daque: Yeah.
220 00:22:24.104 ⇒ 00:22:28.239 Bo Yoon: Can, can you share your emails as well? Just in case
221 00:22:28.860 ⇒ 00:22:37.929 Bo Yoon: so i’ll so for today, i’ll just clean all the all the all the codes that I have all the scripts.
222 00:22:38.950 ⇒ 00:22:43.740 Bo Yoon: and show you a sample of how the the cohort table looks like.
223 00:22:44.790 ⇒ 00:22:47.930 Luke Daque: Cool. Yeah, that’s that’s great. Sounds great.
224 00:22:48.270 ⇒ 00:22:53.150 Bo Yoon: Yeah, this, are you, are you guys working tomorrow as well.
225 00:22:53.380 ⇒ 00:22:54.230 Luke Daque: Yeah.
226 00:22:54.960 ⇒ 00:22:57.340 Bo Yoon: Tomorrow afternoon.
227 00:22:58.220 ⇒ 00:23:05.019 Bo Yoon: so maybe sometime in the afternoon we can. We can meet again, and I can explain to you what my code is.
228 00:23:05.330 ⇒ 00:23:06.010 Luke Daque: Cool.
229 00:23:06.810 ⇒ 00:23:07.290 Nicolas Sucari: Oh, yeah.
230 00:23:07.290 ⇒ 00:23:07.850 Bo Yoon: Okay.
231 00:23:08.800 ⇒ 00:23:12.150 Nicolas Sucari: Same time tomorrow. Does that work for you, Paul?
232 00:23:13.204 ⇒ 00:23:17.030 Bo Yoon: I I can’t do morning tomorrow, but I can do it.
233 00:23:17.030 ⇒ 00:23:18.567 Nicolas Sucari: Hey? Sorry. Yeah.
234 00:23:19.080 ⇒ 00:23:19.710 Bo Yoon: Yeah, it makes sense.
235 00:23:19.710 ⇒ 00:23:20.529 Bo Yoon: I know, that’s okay.
236 00:23:20.530 ⇒ 00:23:24.810 Nicolas Sucari: We are in 3. We are in 3 different time zones. So when you say afternoon, oh.
237 00:23:24.810 ⇒ 00:23:28.680 Nicolas Sucari: for for you, yeah, for any kind of.
238 00:23:28.680 ⇒ 00:23:31.469 Bo Yoon: Located at. Are you based at?
239 00:23:31.870 ⇒ 00:23:32.420 Bo Yoon: Is it.
240 00:23:32.420 ⇒ 00:23:34.960 Nicolas Sucari: I’m in Buenos Aires, Argentina. Yeah.
241 00:23:34.960 ⇒ 00:23:37.900 Bo Yoon: Oh, Argentina, okay, what? What?
242 00:23:37.900 ⇒ 00:23:38.280 Nicolas Sucari: And look.
243 00:23:38.280 ⇒ 00:23:38.810 Bo Yoon: Absolutely.
244 00:23:39.230 ⇒ 00:23:42.480 Luke Daque: I’m in the Philippines, southeast Asia. So.
245 00:23:42.730 ⇒ 00:23:44.599 Bo Yoon: Oh, Philippines, okay.
246 00:23:44.600 ⇒ 00:23:46.400 Luke Daque: We did different time zones.
247 00:23:46.400 ⇒ 00:23:51.254 Bo Yoon: Argentina. I’m pretty sure we’re we’re close in time.
248 00:23:52.418 ⇒ 00:23:56.329 Bo Yoon: Okay, yeah, let’s let’s get a time.
249 00:23:56.330 ⇒ 00:23:59.640 Luke Daque: What time is it? Now? There.
250 00:24:00.038 ⇒ 00:24:04.290 Bo Yoon: In in Pacific standard time. This is here is 10.
251 00:24:04.290 ⇒ 00:24:05.290 Nicolas Sucari: Almost 11.
252 00:24:05.290 ⇒ 00:24:06.859 Bo Yoon: Am, yeah, almost 11.
253 00:24:06.860 ⇒ 00:24:07.800 Nicolas Sucari: Posted online.
254 00:24:07.800 ⇒ 00:24:12.459 Luke Daque: Cool. Yeah. And here it’s like 2 50 am. Almost 3 Pm.
255 00:24:12.460 ⇒ 00:24:13.080 Bo Yoon: Am.
256 00:24:13.370 ⇒ 00:24:13.930 Bo Yoon: Okay.
257 00:24:13.930 ⇒ 00:24:18.040 Bo Yoon: Okay, okay, that’s fine timing for you.
258 00:24:18.210 ⇒ 00:24:18.620 Luke Daque: That’s.
259 00:24:19.410 ⇒ 00:24:22.390 Luke Daque: I’m I’m used to working this hour.
260 00:24:23.860 ⇒ 00:24:34.659 Bo Yoon: Hmm, okay. Oh, and can you also share me? Like a like documentations of of what the Dvt does like like a website that I can take a look at.
261 00:24:34.920 ⇒ 00:24:37.830 Luke Daque: Yeah, sure, I think Nico or.
262 00:24:37.830 ⇒ 00:24:38.210 Nicolas Sucari: Yeah.
263 00:24:38.210 ⇒ 00:24:41.239 Luke Daque: In the chat you can. You can take a look at that.
264 00:24:41.240 ⇒ 00:24:44.340 Bo Yoon: Oh, oh, DVD, oh, get TV! Oh, got it!
265 00:24:44.660 ⇒ 00:24:50.589 Luke Daque: And it has. Yeah, that’s pretty decent documentation there as well. You can check out. Yeah.
266 00:24:51.010 ⇒ 00:24:51.240 Bo Yoon: Yeah.
267 00:24:51.320 ⇒ 00:24:53.610 Luke Daque: Like the learning page. For example.
268 00:24:54.060 ⇒ 00:24:56.050 Bo Yoon: And then there’s sharing.
269 00:24:56.770 ⇒ 00:24:58.810 Luke Daque: Learn. And okay.
270 00:24:59.970 ⇒ 00:25:04.929 Bo Yoon: Yeah. So I’ll get this number tomorrow morning we can meet.
271 00:25:06.730 ⇒ 00:25:15.551 Bo Yoon: So how? When is look, are you? Are you okay with with morning your time
272 00:25:16.390 ⇒ 00:25:20.380 Luke Daque: Like? What? What? So afternoon? New York time, right?
273 00:25:21.280 ⇒ 00:25:26.459 Bo Yoon: Yeah, afternoon. My time. I I can. I can always do like late night as well. So
274 00:25:26.850 ⇒ 00:25:34.190 Bo Yoon: let’s let’s schedule a time. Do you guys use anything? The 1st for scheduling right?
275 00:25:34.190 ⇒ 00:25:39.499 Nicolas Sucari: Yes, I can. I can schedule. Let me let me check. I wanna see all of the time zones.
276 00:25:40.130 ⇒ 00:25:47.120 Nicolas Sucari: For you guys, let me share. And we can use this, maybe time, body work, time, body.
277 00:25:49.120 ⇒ 00:25:51.839 Bo Yoon: Yeah, I always use this one to meet.
278 00:25:54.830 ⇒ 00:25:56.835 Nicolas Sucari: Website when I do this. But
279 00:25:57.590 ⇒ 00:26:00.199 Bo Yoon: If you guys have anything else that’ll be great.
280 00:26:01.720 ⇒ 00:26:07.760 Nicolas Sucari: Oh, I mean I can. I can scale. But so you are pacific, right?
281 00:26:08.330 ⇒ 00:26:09.370 Bo Yoon: Yes.
282 00:26:09.670 ⇒ 00:26:12.020 Nicolas Sucari: And Philippines.
283 00:26:13.960 ⇒ 00:26:19.649 Nicolas Sucari: Okay, what will be a good time for you both at any afternoon.
284 00:26:20.350 ⇒ 00:26:26.330 Bo Yoon: So so in my time, anytime after 1230 Pm. Will be will be good.
285 00:26:26.810 ⇒ 00:26:32.440 Nicolas Sucari: Okay. So somewhere between here, starting from here. What do you think?
286 00:26:32.630 ⇒ 00:26:37.699 Nicolas Sucari: 1230, as as early as possible, or later in the day.
287 00:26:38.500 ⇒ 00:26:40.650 Luke Daque: Yeah, maybe like one or 2.
288 00:26:41.160 ⇒ 00:26:42.190 Luke Daque: Okay.
289 00:26:42.514 ⇒ 00:26:44.459 Nicolas Sucari: Are you okay? Both with one.
290 00:26:45.780 ⇒ 00:26:48.440 Bo Yoon: 1 1 o’clock. Yeah, yeah. Sure. That’s
291 00:26:48.600 ⇒ 00:26:51.589 Bo Yoon: if that’s okay with you guys, then, yeah, it’s totally cool.
292 00:26:51.930 ⇒ 00:26:52.810 Luke Daque: Cool. Yeah.
293 00:26:53.010 ⇒ 00:26:55.410 Nicolas Sucari: I’ll send the invite guys. Yeah.
294 00:26:56.260 ⇒ 00:27:03.070 Bo Yoon: Okay? Sure? Then, yeah, I’ll get started on cleaning the data, and I’ll show you my code tomorrow. Then share my code.
295 00:27:03.380 ⇒ 00:27:03.940 Luke Daque: Sounds good.
296 00:27:03.940 ⇒ 00:27:24.879 Nicolas Sucari: Perfect. If you wanna if you wanna send ahead of the meeting your your script in the slack channel, feel free to do that so that we can take a look. Yeah. And then we’ll we’ll see that tomorrow. And we we then look, we can discuss on how we can implement that into DVD and helpful to get this into production easily. Okay.
297 00:27:25.240 ⇒ 00:27:33.471 Luke Daque: Sounds good. Okay, yeah, that’ll be great. Thank you so much. Have a good night. Have a good day.
298 00:27:33.890 ⇒ 00:27:35.680 Bo Yoon: You too. Bye, bye.