Meeting Title: Analytics Engineering Daily Sync Date: 2025-03-06 Meeting participants: Luke Daque, Caio Velasco, Uttam Kumaran
WEBVTT
1 00:01:01.200 ⇒ 00:01:02.130 Caio Velasco: Hey, Luke?
2 00:02:24.020 ⇒ 00:02:25.950 Caio Velasco: Hey, man! How’s it going? Are you there?
3 00:02:30.890 ⇒ 00:02:31.530 Luke Daque: Hello!
4 00:02:32.190 ⇒ 00:02:33.620 Luke Daque: Hi, yes, I can hear you.
5 00:02:33.620 ⇒ 00:02:35.439 Luke Daque: There. You yeah, I was. That was one
6 00:02:35.770 ⇒ 00:02:38.139 Luke Daque: my mic wasn’t on for some reason.
7 00:02:38.550 ⇒ 00:02:40.510 Luke Daque: Yeah, I’m I’m doing well. How are you?
8 00:02:40.890 ⇒ 00:02:41.870 Caio Velasco: Good! Good!
9 00:02:41.980 ⇒ 00:02:44.610 Caio Velasco: Oh, oh, we will as well.
10 00:02:45.360 ⇒ 00:02:46.170 Luke Daque: Nice.
11 00:02:49.480 ⇒ 00:02:51.369 Caio Velasco: Me so far.
12 00:02:52.230 ⇒ 00:02:52.910 Luke Daque: Still have.
13 00:02:53.431 ⇒ 00:02:56.040 Caio Velasco: Yeah. So yesterday I was
14 00:02:57.140 ⇒ 00:03:04.200 Caio Velasco: well checking all the messages and and slack, and and then I started to play around
15 00:03:04.530 ⇒ 00:03:07.820 Caio Velasco: like with chat dpt, and also to do some research.
16 00:03:08.880 ⇒ 00:03:09.280 Luke Daque: Yeah.
17 00:03:09.280 ⇒ 00:03:14.390 Caio Velasco: And it’s interesting that, like this thing that we seem to be trying to do in terms of
18 00:03:15.371 ⇒ 00:03:20.269 Caio Velasco: making the process between engineers and analysts take more.
19 00:03:20.270 ⇒ 00:03:20.760 Luke Daque: Efficient.
20 00:03:21.500 ⇒ 00:03:23.789 Caio Velasco: Somehow more efficient or or better.
21 00:03:24.506 ⇒ 00:03:28.780 Caio Velasco: It’s something that it doesn’t really exist in the market. At least.
22 00:03:28.950 ⇒ 00:03:31.069 Caio Velasco: that’s what Chat did. You were saying.
23 00:03:32.020 ⇒ 00:03:32.660 Luke Daque: Yeah.
24 00:03:33.170 ⇒ 00:03:39.239 Caio Velasco: So it’s interesting because you have semantic layers like cube cube. Js, I don’t know if you heard about
25 00:03:39.510 ⇒ 00:03:40.380 Caio Velasco: yeah.
26 00:03:40.380 ⇒ 00:03:40.760 Luke Daque: Yeah.
27 00:03:40.760 ⇒ 00:03:49.689 Caio Velasco: Other stuff and data catalog. There are many things, but for some reason they are not connected.
28 00:03:50.890 ⇒ 00:03:54.570 Luke Daque: Yeah, it’s probably difficult. It’s like, it’s
29 00:03:55.470 ⇒ 00:04:00.170 Luke Daque: from different tools and stuff like that. So yeah.
30 00:04:02.900 ⇒ 00:04:06.649 Caio Velasco: Baby, hey? How’s it going with tech bit.
31 00:04:08.270 ⇒ 00:04:12.719 Luke Daque: Yeah. Still, having some, it’s it’s pretty complicated like to get the
32 00:04:13.670 ⇒ 00:04:17.589 Luke Daque: like the tokenized revenue token based revenue.
33 00:04:18.600 ⇒ 00:04:23.149 Luke Daque: But yeah, still still working on that part.
34 00:04:23.720 ⇒ 00:04:30.259 Caio Velasco: Yeah. Can you elaborate a bit more? What do they mean by tokenized and revenue?
35 00:04:31.100 ⇒ 00:04:32.430 Luke Daque: Yeah, the like.
36 00:04:32.910 ⇒ 00:04:40.730 Luke Daque: Have you check like, have you tried like a pro AI account, or something like Chat gpt plus, or something?
37 00:04:41.430 ⇒ 00:04:42.249 Luke Daque: No, not yet.
38 00:04:42.250 ⇒ 00:04:44.340 Luke Daque: If if you if you do that
39 00:04:45.358 ⇒ 00:04:47.911 Luke Daque: you can actually have
40 00:04:49.510 ⇒ 00:04:53.850 Luke Daque: what do you call this? You can access the Api right for.
41 00:04:53.850 ⇒ 00:04:54.460 Caio Velasco: Improvement.
42 00:04:54.460 ⇒ 00:05:07.259 Luke Daque: And like, create your own agents or Gpts, or like, or even just do queries using the Api. And if you do that, it’s like token based. So you have this, like every query, is
43 00:05:07.510 ⇒ 00:05:11.220 Luke Daque: X amount of tokens depending on, I guess the
44 00:05:11.670 ⇒ 00:05:13.780 Luke Daque: how long it is or how how
45 00:05:14.010 ⇒ 00:05:18.290 Luke Daque: how many words, the Gpt will respond to you, or something like that.
46 00:05:18.600 ⇒ 00:05:21.459 Luke Daque: So like the payment there is like
47 00:05:22.190 ⇒ 00:05:24.260 Luke Daque: based on the number of tokens.
48 00:05:24.870 ⇒ 00:05:31.569 Luke Daque: So I think it’s this similar to what the stack Blitz is doing. So they have customers
49 00:05:31.690 ⇒ 00:05:38.609 Luke Daque: basically pay them like a monthly price. But that’s there’s
50 00:05:38.950 ⇒ 00:05:43.050 Luke Daque: and they get like a set amount of tokens that they can use.
51 00:05:43.970 ⇒ 00:05:47.699 Luke Daque: But yeah, sometimes they get they reach the limit
52 00:05:48.080 ⇒ 00:05:53.149 Luke Daque: of the tokens, like, let’s say they they’re given a thousand tokens, and then they get
53 00:05:53.320 ⇒ 00:05:57.149 Luke Daque: they are reach the limit even before the month ends.
54 00:05:57.290 ⇒ 00:06:02.189 Luke Daque: So basically they can’t do queries anymore because they don’t have enough tokens.
55 00:06:02.370 ⇒ 00:06:07.930 Luke Daque: and sometimes what they do is they just buy or purchase another
56 00:06:08.050 ⇒ 00:06:17.420 Luke Daque: set of another 1,000 tokens, for example. So that’s like already outside their monthly payment. Right? So
57 00:06:17.790 ⇒ 00:06:20.870 Luke Daque: so, yeah, that’s what I’m trying to figure out.
58 00:06:22.500 ⇒ 00:06:26.870 Luke Daque: Yeah, I I’ve found, like there’s there’s an event table
59 00:06:27.000 ⇒ 00:06:31.130 Luke Daque: that’s related to token limit reached.
60 00:06:31.690 ⇒ 00:06:33.970 Luke Daque: So what I’m trying to figure out is
61 00:06:34.470 ⇒ 00:06:40.960 Luke Daque: if there’s another table, that where the the customer or the user is adding more tokens.
62 00:06:41.180 ⇒ 00:06:46.879 Luke Daque: and maybe I can utilize that to get the revenue for that piece, basically.
63 00:06:48.610 ⇒ 00:06:51.920 Luke Daque: Yeah. And then another challenge is like trying to
64 00:06:52.370 ⇒ 00:06:56.129 Luke Daque: link that event into the actual customer
65 00:06:56.950 ⇒ 00:07:00.250 Luke Daque: based on their customer. Id, just like. The
66 00:07:01.360 ⇒ 00:07:11.090 Luke Daque: the only information that we have in the events is the user id, which is different from the customer. Id so like there’s some complications there. So I’ll have to like, figure out
67 00:07:11.240 ⇒ 00:07:16.339 Luke Daque: that piece as well. So yeah, something like that.
68 00:07:17.660 ⇒ 00:07:19.332 Caio Velasco: Interesting. Very interesting.
69 00:07:20.510 ⇒ 00:07:21.210 Luke Daque: Yeah.
70 00:07:23.260 ⇒ 00:07:31.089 Luke Daque: It’s not a typical yeah, it’s it’s not very typical of that, right? Like that. That kind of.
71 00:07:31.417 ⇒ 00:07:33.709 Caio Velasco: Do. Yeah. Do you feel like you’re
72 00:07:34.100 ⇒ 00:07:37.500 Caio Velasco: the challenge you have now is on the
73 00:07:38.010 ⇒ 00:07:42.919 Caio Velasco: let’s say, MoD data modeling, part meaning modeling
74 00:07:43.560 ⇒ 00:07:47.740 Caio Velasco: from a data point of view after you have access to tables?
75 00:07:47.910 ⇒ 00:08:01.070 Caio Velasco: Or is it a challenge re related to their business itself, like, what? How to define those tokens. In the 1st place, because I mean they built it.
76 00:08:02.090 ⇒ 00:08:07.579 Caio Velasco: All those definitions right like, how would you count a token, or what said? What.
77 00:08:09.290 ⇒ 00:08:12.879 Luke Daque: Yeah, that’s a good question. I don’t know if
78 00:08:14.080 ⇒ 00:08:19.469 Luke Daque: yeah, I don’t know at the moment. Like, if it is.
79 00:08:20.850 ⇒ 00:08:23.879 Luke Daque: yeah, I don’t know if it’s like a business thing.
80 00:08:26.550 ⇒ 00:08:28.210 Luke Daque: Or if it’s
81 00:08:31.340 ⇒ 00:08:36.409 Luke Daque: yeah, if it’s just just data modeling, maybe I just don’t know where to get it. Or like.
82 00:08:37.460 ⇒ 00:08:39.159 Luke Daque: yeah, I’m not sure. At this point.
83 00:08:39.710 ⇒ 00:08:43.420 Caio Velasco: Okay, yeah, looks like it looks like something very interesting.
84 00:08:44.310 ⇒ 00:08:45.060 Luke Daque: Yeah.
85 00:08:47.830 ⇒ 00:08:50.239 Caio Velasco: By the way, is, you can come into this one.
86 00:08:51.675 ⇒ 00:08:56.470 Luke Daque: I don’t know. He did not mention anything in slack.
87 00:08:57.330 ⇒ 00:08:58.330 Luke Daque: Yeah.
88 00:08:58.940 ⇒ 00:09:01.699 Luke Daque: And did he accept this invite? Let me check.
89 00:09:03.850 ⇒ 00:09:07.450 Luke Daque: He did accept the invite, so maybe he should join.
90 00:09:07.590 ⇒ 00:09:12.369 Luke Daque: I wish declined, and I I think he’s out of office.
91 00:09:12.700 ⇒ 00:09:19.440 Luke Daque: and I guess Demilade is just been added to the invite. So I’m not sure if he’s joining.
92 00:09:20.860 ⇒ 00:09:25.800 Luke Daque: Yeah, maybe let’s wait for a bit. Maybe Utah will be joining in time soon.
93 00:09:29.330 ⇒ 00:09:33.850 Caio Velasco: Great, can you? I mean, if you want otherwise? No, no worries at all.
94 00:09:34.060 ⇒ 00:09:34.840 Caio Velasco: Perfect.
95 00:09:35.090 ⇒ 00:09:36.990 Caio Velasco: Share your screen, and then.
96 00:09:37.090 ⇒ 00:09:43.679 Caio Velasco: and show me, like the the basics of the basics of the basics of the this tokenization thing.
97 00:09:44.380 ⇒ 00:09:46.160 Caio Velasco: If you want, don’t, don’t be afraid.
98 00:09:46.160 ⇒ 00:09:46.700 Luke Daque: Yeah.
99 00:09:46.700 ⇒ 00:09:48.350 Caio Velasco: Just because I’m interested.
100 00:09:49.240 ⇒ 00:09:53.329 Luke Daque: Yeah, that’s yes. The way to do that is, maybe we can go through
101 00:09:53.940 ⇒ 00:09:57.050 Luke Daque: the or the stack Blitz site.
102 00:09:57.530 ⇒ 00:10:03.559 Luke Daque: So you can understand. So yeah, let’s let’s try that. Wait. Let me share my screen
103 00:10:04.770 ⇒ 00:10:08.360 Luke Daque: straight to their site to see their plans and stuff.
104 00:10:10.870 ⇒ 00:10:13.550 Luke Daque: I didn’t do that. So can you see my screen now?
105 00:10:14.350 ⇒ 00:10:18.282 Caio Velasco: Yes, and where is it? Where is that location, that beach.
106 00:10:19.435 ⇒ 00:10:19.929 Luke Daque: What?
107 00:10:20.160 ⇒ 00:10:21.480 Luke Daque: Okay?
108 00:10:22.180 ⇒ 00:10:26.660 Luke Daque: Let’s try to say that next in.
109 00:10:27.690 ⇒ 00:10:31.980 Luke Daque: And yeah, so get this one.
110 00:10:32.820 ⇒ 00:10:33.560 Luke Daque: Right?
111 00:10:45.700 ⇒ 00:10:51.690 Luke Daque: So this is their current looks like, this is the pricing right? There’s personal, which is free
112 00:10:52.090 ⇒ 00:10:55.209 Luke Daque: pro teams and something.
113 00:10:56.580 ⇒ 00:11:03.680 Luke Daque: And this and I guess if we if we go to the pro which is build per month.
114 00:11:04.250 ⇒ 00:11:06.970 Luke Daque: You can just join just so that you know.
115 00:11:08.860 ⇒ 00:11:09.280 Uttam Kumaran: Hey, guys.
116 00:11:10.120 ⇒ 00:11:10.830 Caio Velasco: Hello!
117 00:11:10.830 ⇒ 00:11:12.009 Luke Daque: Oh, Hi! Hi Utah!
118 00:11:12.400 ⇒ 00:11:13.590 Uttam Kumaran: Hey? How’s it going.
119 00:11:14.870 ⇒ 00:11:25.520 Luke Daque: Doing? Well, we’re just trying to discuss about like the tokenization of stack bits. I, yeah, like, Kyle wanted to understand how it’s being.
120 00:11:27.710 ⇒ 00:11:31.380 Luke Daque: Yeah, what? How it? What it means, basically
121 00:11:32.270 ⇒ 00:11:37.080 Luke Daque: like how Mitch explained it? Right? Because they have, like a monthly
122 00:11:39.150 ⇒ 00:11:46.360 Luke Daque: thing for the pro users and teams will be 29 per month or something. This can be the monthly, or can be annually.
123 00:11:46.770 ⇒ 00:11:53.119 Luke Daque: But then there’s a limit, even if it’s like $18 per month. There’s actually a limit to the number of tokens.
124 00:11:54.010 ⇒ 00:12:00.509 Luke Daque: And so yeah, like, Kyle wanted to understand more about that tokenization.
125 00:12:00.875 ⇒ 00:12:01.240 Uttam Kumaran: Okay.
126 00:12:02.240 ⇒ 00:12:02.760 Caio Velasco: Yeah, just.
127 00:12:02.760 ⇒ 00:12:03.230 Luke Daque: That’s.
128 00:12:03.230 ⇒ 00:12:06.519 Caio Velasco: Curious, because it it does sound quite interesting. How they do.
129 00:12:06.520 ⇒ 00:12:07.540 Uttam Kumaran: Yeah, yeah.
130 00:12:08.710 ⇒ 00:12:10.090 Luke Daque: Yeah, it’s cool the way they do it.
131 00:12:11.530 ⇒ 00:12:12.440 Luke Daque: Yeah.
132 00:12:13.620 ⇒ 00:12:15.850 Uttam Kumaran: You may have to just sign up for an account like.
133 00:12:16.540 ⇒ 00:12:18.029 Luke Daque: Yeah, I guess so.
134 00:12:22.050 ⇒ 00:12:27.600 Uttam Kumaran: Or sign up? I mean, do we have a brain like.
135 00:12:27.710 ⇒ 00:12:31.250 Uttam Kumaran: can you? Do you have to sign up with Google? Or can you just sign up with like a.
136 00:12:31.250 ⇒ 00:12:34.100 Luke Daque: You can sign up using github looks like.
137 00:12:34.640 ⇒ 00:12:35.180 Uttam Kumaran: Okay.
138 00:12:35.180 ⇒ 00:12:35.870 Luke Daque: Right.
139 00:12:36.160 ⇒ 00:12:38.389 Uttam Kumaran: Or you can sign up with an email password.
140 00:12:39.400 ⇒ 00:12:42.849 Luke Daque: Yeah. And then to use the stack with Gmail.
141 00:12:43.620 ⇒ 00:12:46.079 Uttam Kumaran: Is there a brain forge at Stackblitz?
142 00:12:49.700 ⇒ 00:12:51.509 Luke Daque: Yeah, wait a minute. Second. Let me.
143 00:12:51.510 ⇒ 00:12:54.950 Uttam Kumaran: Or there’s a i have a stack Blitz email. Right? Yeah.
144 00:12:55.250 ⇒ 00:12:57.240 Luke Daque: Yeah, we can use that. Then.
145 00:12:57.240 ⇒ 00:13:02.060 Uttam Kumaran: Yeah, maybe just create an account there. And that way I can even ask him to give us like an upgrade. And.
146 00:13:02.370 ⇒ 00:13:03.219 Luke Daque: We wanna test
147 00:13:14.480 ⇒ 00:13:15.980 Luke Daque: this one? Right?
148 00:13:26.980 ⇒ 00:13:28.780 Luke Daque: I’ll use the same password.
149 00:13:38.180 ⇒ 00:13:39.749 Luke Daque: Oh, we need to create them.
150 00:13:44.630 ⇒ 00:13:47.510 Luke Daque: Yeah, I can do this. Maybe after the meeting. Just to like.
151 00:13:47.510 ⇒ 00:13:47.915 Uttam Kumaran: Okay.
152 00:13:53.450 ⇒ 00:13:56.239 Uttam Kumaran: cool. Yeah. The only thing I’m working on is,
153 00:13:57.510 ⇒ 00:13:59.685 Uttam Kumaran: So yeah, we had
154 00:14:00.840 ⇒ 00:14:05.410 Uttam Kumaran: Steven started this week just on Pm stuff for Javi.
155 00:14:05.610 ⇒ 00:14:09.040 Uttam Kumaran: So we’re working on the larger roadmap there.
156 00:14:11.410 ⇒ 00:14:18.920 Uttam Kumaran: And then, yeah, I’m I’m also we. We onboarded another Pm. Amber, who’s coming on to take over
157 00:14:19.030 ⇒ 00:14:24.910 Uttam Kumaran: ABC home, and then slowly they’ll start to
158 00:14:25.170 ⇒ 00:14:29.739 Uttam Kumaran: probably split the the rest of the clients as well, and start to build out
159 00:14:30.200 ⇒ 00:14:35.670 Uttam Kumaran: roadmaps everywhere. The other piece I’m I’m starting to work on is basically
160 00:14:36.190 ⇒ 00:14:38.079 Uttam Kumaran: starting to make sure there’s at least
161 00:14:40.440 ⇒ 00:14:43.770 Uttam Kumaran: you know, probably 2 aes on every client.
162 00:14:43.920 ⇒ 00:14:47.830 Uttam Kumaran: But for the most part, like, I think, every client
163 00:14:48.368 ⇒ 00:14:53.790 Uttam Kumaran: like some of the clients. It’s they’re small, for example, like small, even in terms of
164 00:14:54.080 ⇒ 00:14:57.500 Uttam Kumaran: what we are, what we are charging the client
165 00:14:57.900 ⇒ 00:14:59.340 Uttam Kumaran: like, I think, in the last.
166 00:14:59.860 ⇒ 00:15:17.000 Uttam Kumaran: In the last 2 weeks we had to do a lot for Javi. But we definitely like we spent more fixing that than we’re gonna get paid on that one which is fine, which is fine. However, there we sort of have 2 buckets of clients. We have.
167 00:15:17.210 ⇒ 00:15:18.820 Uttam Kumaran: We have like 2 plans.
168 00:15:19.240 ⇒ 00:15:26.469 Uttam Kumaran: And so for the smaller plan, typically, we spend anywhere from like 10 to 20 HA week.
169 00:15:27.210 ⇒ 00:15:31.219 Uttam Kumaran: for the larger plan. It ends up being more like.
170 00:15:31.590 ⇒ 00:15:34.119 Uttam Kumaran: you know, like anywhere from 20 to 40.
171 00:15:34.390 ⇒ 00:15:37.159 Uttam Kumaran: That’s that plan is where Eden is.
172 00:15:37.780 ⇒ 00:15:42.410 Luke Daque: And then we have one more client that is probably gonna start at that same.
173 00:15:42.920 ⇒ 00:15:44.539 Uttam Kumaran: At that similar level.
174 00:15:46.440 ⇒ 00:15:54.769 Uttam Kumaran: so. But the but the one thing I you know with this team is, I want to make sure there’s at least 2 aes on every like one person to
175 00:15:54.890 ⇒ 00:15:59.100 Uttam Kumaran: to like one person to
176 00:15:59.200 ⇒ 00:16:03.809 Uttam Kumaran: look at code, one person to to write code, and someone to review, and then also, in case
177 00:16:03.980 ⇒ 00:16:11.262 Uttam Kumaran: we’re out, you can sort of begin to hand that off.
178 00:16:12.300 ⇒ 00:16:12.820 Luke Daque: Yeah.
179 00:16:12.820 ⇒ 00:16:20.350 Uttam Kumaran: Like if you’re out of office or something, there’s like some redundancy. So that’s sort of what I’m working on. We’re working on like a larger org org chart right now.
180 00:16:20.580 ⇒ 00:16:31.149 Uttam Kumaran: So I think I’m gonna try and present something by tomorrow to everybody. But if you see me adding people to to channels. That’s sort of what I’m what I’m working on.
181 00:16:34.570 ⇒ 00:16:35.300 Luke Daque: Okay.
182 00:16:36.500 ⇒ 00:16:37.460 Luke Daque: Cool.
183 00:16:37.690 ⇒ 00:16:41.380 Uttam Kumaran: That way. There’s there’s sort of 2 people with context on on every
184 00:16:41.980 ⇒ 00:16:45.820 Uttam Kumaran: yeah repo for each client, you know, or at least some context.
185 00:16:46.760 ⇒ 00:16:54.940 Luke Daque: Yeah, it’s pretty difficult as well, for like new joiners like Kyle, for example, where, like, Jabby, coffee’s already been built like.
186 00:16:55.500 ⇒ 00:16:56.290 Luke Daque: yeah.
187 00:16:56.290 ⇒ 00:17:05.040 Luke Daque: So it’s like he doesn’t really have context how the models were built, or what those like, what the gorgeous messages were. So it’s like
188 00:17:05.270 ⇒ 00:17:13.039 Luke Daque: he already like we already did that before. And then he’s trying to like, we spend time trying to understand that before
189 00:17:13.550 ⇒ 00:17:19.229 Luke Daque: now. He’s yeah trying to understand it again. So it’s like duplicating the work right or something.
190 00:17:19.230 ⇒ 00:17:20.020 Uttam Kumaran: Yeah, and that’s.
191 00:17:20.020 ⇒ 00:17:20.509 Luke Daque: Just because.
192 00:17:20.510 ⇒ 00:17:29.369 Uttam Kumaran: We didn’t. Yeah, I think that’ll change now, as like we build back up that redundancy. And then for new clients, we’ll always start with
193 00:17:29.700 ⇒ 00:17:31.140 Uttam Kumaran: with 2 people.
194 00:17:31.550 ⇒ 00:17:31.960 Luke Daque: And.
195 00:17:33.140 ⇒ 00:17:40.830 Uttam Kumaran: I mean, we’ll see, I think, as we grow we’ll sort of as even as a team, and and individually we will see
196 00:17:41.150 ⇒ 00:17:47.540 Uttam Kumaran: like, I just want there to be more people reviewing code than just me.
197 00:17:47.730 ⇒ 00:17:55.560 Uttam Kumaran: And so I also want there to be like, Okay, hey, I’m gonna be out, can you just like handle anything that comes up for
198 00:17:55.810 ⇒ 00:18:01.870 Uttam Kumaran: Eden right right now? That’s me like a waste was out. And I was like, Okay, I’m gonna I gotta handle
199 00:18:02.150 ⇒ 00:18:05.400 Uttam Kumaran: basically everything for Eden. And I don’t really like I
200 00:18:05.680 ⇒ 00:18:11.619 Uttam Kumaran: I know, I know, like a good amount. But it takes a lot for me to like refresh on like what I’m doing there.
201 00:18:12.600 ⇒ 00:18:13.330 Luke Daque: Yeah.
202 00:18:13.910 ⇒ 00:18:15.479 Uttam Kumaran: So I think between
203 00:18:15.680 ⇒ 00:18:23.520 Uttam Kumaran: yeah, between this crew and demalade, I think we will have basically redundancy across most clients. So which is good.
204 00:18:25.530 ⇒ 00:18:30.859 Luke Daque: Yeah, like I, I might also be out for a few weeks on
205 00:18:30.970 ⇒ 00:18:35.280 Luke Daque: July to August, because I’ll be traveling there to
206 00:18:36.570 ⇒ 00:18:39.599 Luke Daque: yeah, yes, maybe we can even put on there.
207 00:18:39.940 ⇒ 00:18:41.000 Uttam Kumaran: Oh, nice!
208 00:18:42.250 ⇒ 00:18:47.660 Luke Daque: I’ll be. Yeah. I think I mentioned this before. I’d be like joining I mean, I was invited to a wedding.
209 00:18:48.984 ⇒ 00:18:50.719 Luke Daque: Yeah, my cousin’s wedding.
210 00:18:51.160 ⇒ 00:18:53.749 Luke Daque: Yeah dude in New New Jersey, or something.
211 00:18:53.750 ⇒ 00:18:55.270 Uttam Kumaran: Tell me when you’re going to be there.
212 00:18:55.660 ⇒ 00:18:58.779 Luke Daque: Sure we haven’t. Don’t have a
213 00:18:59.380 ⇒ 00:19:04.130 Luke Daque: ticket yet, but somewhere around July end of July to
214 00:19:04.450 ⇒ 00:19:06.269 Luke Daque: 1st week of August, I guess.
215 00:19:09.100 ⇒ 00:19:11.770 Uttam Kumaran: Amazing. Well, yeah, I mean, I’ll come see you in New York.
216 00:19:12.300 ⇒ 00:19:13.650 Luke Daque: Sure that’d be great.
217 00:19:18.600 ⇒ 00:19:21.730 Luke Daque: Yeah, I’ll be in New York for a couple of days or so.
218 00:19:23.540 ⇒ 00:19:24.080 Uttam Kumaran: Dude.
219 00:19:25.140 ⇒ 00:19:34.920 Uttam Kumaran: Yeah, I mean, you’re you’re definitely welcome to come to Austin, but I know it’s a you’re going to be with family, so I’ll come see you then. Robert’s there in New York, so we can all hang out.
220 00:19:35.370 ⇒ 00:19:36.670 Luke Daque: Noise. Yeah.
221 00:19:38.790 ⇒ 00:19:45.449 Uttam Kumaran: Yeah, I don’t know. I want to like it’s we’ll see this year if I’m able to do this. But I want to try to get everyone together.
222 00:19:45.590 ⇒ 00:19:52.549 Uttam Kumaran: I don’t know if we can. We’re gonna be able to afford that this year. I think I’m gonna be able to afford to go see everyone at least once.
223 00:19:53.440 ⇒ 00:19:55.320 Uttam Kumaran: Like I saw Nico.
224 00:19:55.810 ⇒ 00:20:00.670 Uttam Kumaran: I think I’m gonna try to. I mean, I was just talking to Robert. I think I may try to go to
225 00:20:01.210 ⇒ 00:20:08.500 Uttam Kumaran: Manila sometime in the winter, because he’s gonna be he’s gonna be in.
226 00:20:09.360 ⇒ 00:20:19.180 Luke Daque: Yeah, especially. There’s a lot of us, I think right. And Casey, the Ae, the AI team. There’s a lot of people in Manila.
227 00:20:19.180 ⇒ 00:20:27.729 Uttam Kumaran: He’s gonna be in. He’s gonna be in Hong Kong for a wedding. And I was like, Okay, maybe we should both go and see everybody.
228 00:20:27.990 ⇒ 00:20:28.720 Luke Daque: Yeah.
229 00:20:31.290 ⇒ 00:20:34.690 Uttam Kumaran: And then I’ll stop by in Spain while I’m all on my way.
230 00:20:35.330 ⇒ 00:20:36.170 Caio Velasco: Right.
231 00:20:39.237 ⇒ 00:20:45.592 Uttam Kumaran: Yeah, yeah. And I. So I guess a couple. So I guess maybe let’s talk about
232 00:20:47.150 ⇒ 00:20:52.540 Uttam Kumaran: where each of the each of the clients are. So for Java. Yeah, we
233 00:20:52.700 ⇒ 00:20:58.139 Uttam Kumaran: we pushed out a lot of stuff. And so the the discussions are going much better. Now.
234 00:20:58.410 ⇒ 00:21:05.410 Uttam Kumaran: one thing that we’re working on. And I think, Kai, you’re sort of in this like limbo state until we get the next roadmap.
235 00:21:05.860 ⇒ 00:21:11.019 Uttam Kumaran: which is gonna probably take another like couple days
236 00:21:11.733 ⇒ 00:21:15.440 Uttam Kumaran: on like what the next core objectives are.
237 00:21:16.793 ⇒ 00:21:30.700 Uttam Kumaran: In the meantime, though, I think we certainly need, like the documentation process, ae analyst handoff to work for this client like I think we have the best shot.
238 00:21:30.820 ⇒ 00:21:37.900 Uttam Kumaran: because we have this moment where there’s not a lot of stuff we have the best shot at making it happen for Javi.
239 00:21:38.666 ⇒ 00:21:43.940 Uttam Kumaran: I know I didn’t get a chance to reply to your message yesterday, but
240 00:21:45.400 ⇒ 00:21:47.830 Uttam Kumaran: I think the focus for Javi at least
241 00:21:48.331 ⇒ 00:21:53.789 Uttam Kumaran: until early next week. Continue to. For you continues to just remain like, okay, can we?
242 00:21:54.340 ⇒ 00:22:01.590 Uttam Kumaran: Can you start to just consume and become like and own the code base and sort of also own the process of
243 00:22:02.000 ⇒ 00:22:06.050 Uttam Kumaran: okay, how are requirements getting passed down?
244 00:22:06.200 ⇒ 00:22:13.099 Uttam Kumaran: We will have a Pm. For you on this account, though probably by the end of next week.
245 00:22:13.410 ⇒ 00:22:15.119 Uttam Kumaran: if not mid next week.
246 00:22:15.250 ⇒ 00:22:20.599 Uttam Kumaran: which will take all the pressure of like what’s coming up next. What’s working on every day
247 00:22:20.920 ⇒ 00:22:25.079 Uttam Kumaran: that’s going to get solved. But it’s going to take a couple more days. So.
248 00:22:25.080 ⇒ 00:22:25.959 Caio Velasco: You know we’re
249 00:22:26.280 ⇒ 00:22:32.509 Caio Velasco: cool, cool. No, that that looks that looks good to me. And yeah, I got a chance to learn a bit more about
250 00:22:32.690 ⇒ 00:22:36.399 Caio Velasco: stuff that has been done before I started.
251 00:22:37.530 ⇒ 00:22:45.629 Caio Velasco: I can even not, I mean, focus on the on the Amazon source. We then.
252 00:22:45.630 ⇒ 00:22:46.430 Uttam Kumaran: Yeah.
253 00:22:46.430 ⇒ 00:22:52.420 Caio Velasco: You know. Get the hands dirty more in in the financial stuff.
254 00:22:52.560 ⇒ 00:23:04.539 Caio Velasco: Cool? And then start learning. Yeah. See if I can with time on the the process, and know where things are and understand how the the modeling was done. I think that’s important.
255 00:23:05.590 ⇒ 00:23:07.707 Uttam Kumaran: Yeah. Another piece, you know, is
256 00:23:09.190 ⇒ 00:23:12.700 Uttam Kumaran: We’ve modeled Amazon before once.
257 00:23:12.950 ⇒ 00:23:17.940 Uttam Kumaran: So I’m also trying to think. And maybe again, like, this is where my time
258 00:23:18.380 ⇒ 00:23:37.120 Uttam Kumaran: once I get time to sort of see every client, and I’m going to start to build this out is like when we take on new clients. If, how do we build up source level documentation, for example, like we had to model fees again for Amazon, and we already did a once for pool parts.
259 00:23:37.240 ⇒ 00:23:41.819 Uttam Kumaran: What what do we learn right? And how do we make that? How do we make that
260 00:23:42.100 ⇒ 00:23:44.190 Uttam Kumaran: clear? So that
261 00:23:44.630 ⇒ 00:23:51.260 Uttam Kumaran: the team has a place where it’s like, oh, anytime we touch Amazon. Let me go. Check what we’ve done already for Amazon, you know.
262 00:23:53.486 ⇒ 00:24:00.060 Caio Velasco: And but can you give me just a an example so that I can picture like a a situation.
263 00:24:00.470 ⇒ 00:24:06.400 Uttam Kumaran: Yeah. So here’s a good example, like, when you sell on Amazon, they take a fee
264 00:24:06.510 ⇒ 00:24:12.219 Uttam Kumaran: right? And there’s 2 types of selling on Amazon. There’s you can fulfill the order meaning
265 00:24:12.630 ⇒ 00:24:20.499 Uttam Kumaran: you just sell it on Amazon. You get the order, slip you have to package and send the product. There’s also fulfilled by Amazon.
266 00:24:20.750 ⇒ 00:24:28.330 Uttam Kumaran: which means you send them like a. You send them like a hundred 1,000 like units.
267 00:24:28.440 ⇒ 00:24:35.630 Uttam Kumaran: They store it, pack it and ship it, but they charge you a higher fee in order to do that.
268 00:24:36.610 ⇒ 00:24:40.710 Uttam Kumaran: that what that helps with is that they they have really good shipping rates
269 00:24:41.130 ⇒ 00:25:06.409 Uttam Kumaran: like they’ll handle all that stuff you don’t have to hire for that stuff, but they charge a higher rate for that. And so these fees are like the modeling for them is a little bit complicated because the tables are. It’s not really clear like where they are, and for pool parts we spent like a few weeks going, and like I don’t know. We spend even longer. I think probably Ryan, like finding these fees understanding.
270 00:25:06.410 ⇒ 00:25:06.910 Luke Daque: Yeah.
271 00:25:06.910 ⇒ 00:25:09.470 Uttam Kumaran: What all the fee types are.
272 00:25:09.660 ⇒ 00:25:16.870 Uttam Kumaran: How do actually, what table they’re in? And then how does that table join to the, to the transactions
273 00:25:17.020 ⇒ 00:25:22.010 Uttam Kumaran: like it’s, it’s actually not the concept of like, okay, what is a fee? It’s just like the
274 00:25:22.120 ⇒ 00:25:31.510 Uttam Kumaran: there’s just like annoying things about like how to where to get that from, and the naming conventions, and there was like the documentation sucked. So we figured it out.
275 00:25:31.700 ⇒ 00:25:35.045 Uttam Kumaran: We have to do the same thing for for Javi, though.
276 00:25:36.360 ⇒ 00:25:43.829 Uttam Kumaran: And so I think I mean I’m I feel like we. I think it’s there again. I have to go remind myself. But
277 00:25:44.480 ⇒ 00:25:49.460 Uttam Kumaran: and I think we did use a lot of what we learned from the last client there. But that sort of stuff that
278 00:25:49.570 ⇒ 00:25:52.680 Uttam Kumaran: would be great to have that in notion, which is just like.
279 00:25:53.290 ⇒ 00:25:58.970 Uttam Kumaran: Okay, if you’re working with Amazon on a client, here are like 10 things to be aware of or like.
280 00:25:59.140 ⇒ 00:26:06.839 Uttam Kumaran: Here’s a here’s a template database to get started, you know. That sort of stuff.
281 00:26:07.690 ⇒ 00:26:11.379 Caio Velasco: Oh, got it. Okay, I can start from that.
282 00:26:11.590 ⇒ 00:26:12.090 Caio Velasco: Okay.
283 00:26:12.090 ⇒ 00:26:12.850 Uttam Kumaran: Yeah.
284 00:26:17.920 ⇒ 00:26:27.269 Uttam Kumaran: So I think maybe just locking in on. On continuing to document Javi is helpful. I’ve also started to add, I added you to the pool parts.
285 00:26:28.075 ⇒ 00:26:29.800 Uttam Kumaran: Channel as well
286 00:26:30.320 ⇒ 00:26:35.669 Uttam Kumaran: right now. I think it’s just it right now that so that’s 1 of our oldest clients.
287 00:26:36.281 ⇒ 00:26:56.479 Uttam Kumaran: and Ryan. It was pretty previously me. And then it’s been Ryan, and I wanted to also include you there as sort of redundancy, and also there in a moment, where there’s not a lot of modeling going on. So again, a really good opportunity for us to do some really great documentation.
288 00:26:56.780 ⇒ 00:27:02.209 Uttam Kumaran: I mean, this is where it’s like I’m sort of trying to, I mean, I guess I’ll ask you.
289 00:27:02.610 ⇒ 00:27:19.040 Uttam Kumaran: There’s definitely like stuff. I mean, I don’t know. This is where it’s like. It depends on the type of person you are. If you’re for me, I think, like, okay, if you’re in sort of documentation world. And you’re thinking about systems. Let me give you another client where like, that’s sort of the space they’re in that way.
290 00:27:19.320 ⇒ 00:27:23.640 Uttam Kumaran: That way you can start to build these systems that we can reuse.
291 00:27:23.750 ⇒ 00:27:25.239 Uttam Kumaran: However, these aren’t like
292 00:27:25.550 ⇒ 00:27:34.709 Uttam Kumaran: action packed tasks like they’re not like, you know. I don’t know you don’t. You kind of get what I mean. So you tell me, like, there’s other clients that have
293 00:27:35.100 ⇒ 00:27:42.569 Uttam Kumaran: others like actual modeling. But this is really really important. And I think, while you’re thinking about documentation.
294 00:27:42.740 ⇒ 00:27:49.400 Uttam Kumaran: it would be helpful to do that across like 2 clients and start to, you know. Use your time that way.
295 00:27:50.460 ⇒ 00:27:54.360 Caio Velasco: No, I can get. I can definitely do that and actually, I think that
296 00:27:54.710 ⇒ 00:28:00.829 Caio Velasco: I don’t know why. But even even though I like to do the modeling. And for sure that that’s something that
297 00:28:00.930 ⇒ 00:28:05.810 Caio Velasco: I was be doing as a data engineer. I’m a data engineer.
298 00:28:06.619 ⇒ 00:28:11.709 Caio Velasco: Doc, this documentation is actually how I feel that I’m
299 00:28:11.920 ⇒ 00:28:22.460 Caio Velasco: really learning about how the client works. And then when you get the data modeling part well, at least I assume the thing will be easier and more efficient.
300 00:28:23.317 ⇒ 00:28:28.449 Caio Velasco: So yes, that’s for me. I I would always start from there.
301 00:28:29.060 ⇒ 00:28:29.580 Caio Velasco: Okay, great.
302 00:28:29.580 ⇒ 00:28:29.989 Uttam Kumaran: It’s okay.
303 00:28:29.990 ⇒ 00:28:30.540 Uttam Kumaran: Okay.
304 00:28:30.790 ⇒ 00:28:47.460 Uttam Kumaran: okay, that’s great. I mean, I sort of like to ask, like, there’s things I can do. And there’s also things people like to do. But I do think that for for pool parts also, the code base is very mature meaning we haven’t made a lot of changes recently. But documentation is like
305 00:28:48.030 ⇒ 00:28:57.520 Uttam Kumaran: nonexistent would be an understatement. And also it’s more. It’s it’s like, it’s probably like
306 00:28:58.050 ⇒ 00:29:12.180 Uttam Kumaran: it’s probably 100% more complicated than Javi, because we have a lot of manual and like really like ugly sources that we’re pulling from like shipping providers and
307 00:29:12.820 ⇒ 00:29:30.740 Uttam Kumaran: like random emails that we’re getting Csvs from. There’s a lot, but they are going to grow. And so I want to get the documentation done, because otherwise there’s no way that people can start on working on them. A lot of that knowledge is in my brain.
308 00:29:31.460 ⇒ 00:29:38.670 Uttam Kumaran: So it would be helpful, I mean. But so so tell me like how how you think about tackling the documentation like
309 00:29:38.940 ⇒ 00:29:48.019 Uttam Kumaran: or like basically this task as a whole. And how can I help break this down into like phases? Or, for example, if we were gonna create tickets around this like
310 00:29:48.200 ⇒ 00:29:57.659 Uttam Kumaran: my initial thought was to just create like a a broad spike ticket. Typically, that’s just like, okay, go, take a look at it. But do you? Do you you think about
311 00:29:57.960 ⇒ 00:30:00.919 Uttam Kumaran: like, I think what would be helpful is like, how do you think about
312 00:30:01.070 ⇒ 00:30:04.180 Uttam Kumaran: going from 0 to a hundred percent on
313 00:30:04.390 ⇒ 00:30:09.869 Uttam Kumaran: documentation understanding on any client? And can we break that into tasks like, do you have a sense of that?
314 00:30:10.590 ⇒ 00:30:33.970 Caio Velasco: Yeah, I think. Well, I always look at everything we do here as like a pipeline. So always from upstream to downstream, and from sources to to dashboards, so I don’t know if I would. If I were to do this with Job, you know I would probably start from from the beginning like, what are the sources? Are they there? What is happening there? Okay? And then.
315 00:30:34.740 ⇒ 00:30:46.769 Caio Velasco: if if we can do a pair of notion and and spreadsheet notion would be more like guiding you through reading what is happening. I think that’s usually the the the
316 00:30:47.470 ⇒ 00:30:55.629 Caio Velasco: the easier, the easier way to the easiest way to to understand what is happening like with a client, or with with the documentation itself.
317 00:30:55.910 ⇒ 00:31:00.609 Caio Velasco: with the document itself. And then, if you, if I go to the source. And I list
318 00:31:00.940 ⇒ 00:31:30.899 Caio Velasco: some some of the sources. Then, yeah, I want to check check one of the sources like Amazon, or Gorgeous or something. Then I have to put the link to a spreadsheet, and the spreadsheet has also to be part of the work. So you go to the spreadsheet and you say, okay, these are the the kinds of definitions for for cogs, for margin, for whatever this is how you can find in the source. Like, I’m mapping like, basically, I’m mapping. The spreadsheet would be there to map stuff from
319 00:31:31.190 ⇒ 00:31:39.104 Caio Velasco: a column in the source. To what metric is. Is it? Pointing to to the dashboard. Now the thing is
320 00:31:40.660 ⇒ 00:31:50.340 Caio Velasco: If the dashboard that that’s that’s exactly my my question before chat. If if the dashboard changes, or if a metric changes.
321 00:31:50.570 ⇒ 00:31:53.840 Caio Velasco: Does it mean that we lost all the the work
322 00:31:54.080 ⇒ 00:31:56.980 Caio Velasco: cause? Because it’s a mapping situation, right?
323 00:31:57.130 ⇒ 00:32:03.670 Caio Velasco: So that’s how I, where I get a bit stuck in how to proceed with this. But yeah, documentation would go through. So.
324 00:32:03.670 ⇒ 00:32:09.329 Uttam Kumaran: Wait. I guess. I guess. Like, yeah, let’s talk about that. So you’re talking more about like change management.
325 00:32:13.530 ⇒ 00:32:14.819 Caio Velasco: What do you mean by changing.
326 00:32:15.160 ⇒ 00:32:26.390 Uttam Kumaran: Meaning like, let’s say we go make a definition change. Are you talking about like, okay, what downstream like, what steps are there after I change a piece of logic. Okay, okay.
327 00:32:26.390 ⇒ 00:32:46.030 Caio Velasco: Yes, yes, no, perfect, because that’s also let’s say that we are building gorges. And and one of the lines of our spreadsheet is mapping messages to the macro metric on the dashboard. But then, you say, like, Hey, we were using messages. But now we have to use. I know tickets, whatever
328 00:32:47.160 ⇒ 00:32:51.579 Caio Velasco: you know. So how would that be? Because, yeah, we will have to change over there.
329 00:32:51.930 ⇒ 00:32:58.210 Uttam Kumaran: Well, yeah, I think at the moment it’s not gonna be dynamic. But yeah, I can
330 00:32:58.210 ⇒ 00:33:11.239 Uttam Kumaran: like, if I get some time I’ll go make that happen for us. But I think we just have to. We have to. The Aes have to be tasked to update it. So there’s probably 2 opportunities. One is.
331 00:33:11.710 ⇒ 00:33:16.180 Uttam Kumaran: there’s 2. There’s 2 ways we could do this either, for every Pr you make.
332 00:33:16.310 ⇒ 00:33:19.519 Uttam Kumaran: And and this is gonna slow things down is
333 00:33:19.660 ⇒ 00:33:23.970 Uttam Kumaran: you have to go. Make sure that documentation is updated right?
334 00:33:24.280 ⇒ 00:33:24.820 Caio Velasco: Yep.
335 00:33:24.820 ⇒ 00:33:26.280 Uttam Kumaran: The second piece
336 00:33:26.510 ⇒ 00:33:33.169 Uttam Kumaran: we could do is we can push Prs, and at the end of the week we just do tech debt every week.
337 00:33:33.862 ⇒ 00:33:44.189 Uttam Kumaran: Where we go update documentation, basically where we look at all the Prs that were pushed, we make sure that new columns are added into the documentation.
338 00:33:44.310 ⇒ 00:33:48.589 Uttam Kumaran: We make sure that the new metrics are there. New sources are there.
339 00:33:49.170 ⇒ 00:33:52.100 Uttam Kumaran: and we spend an hour as a team.
340 00:33:52.520 ⇒ 00:33:55.660 Uttam Kumaran: or even longer, like just doing that.
341 00:33:57.960 ⇒ 00:34:02.069 Caio Velasco: No, I agree, and I think we even can can use well, the AI thing.
342 00:34:02.070 ⇒ 00:34:04.260 Uttam Kumaran: We can use the AI, yeah.
343 00:34:04.260 ⇒ 00:34:10.729 Caio Velasco: To see like well, what happened in the last few hours, and give me an update of all the tables and columns, and then.
344 00:34:10.739 ⇒ 00:34:11.219 Uttam Kumaran: What I’m.
345 00:34:11.219 ⇒ 00:34:11.899 Caio Velasco: Especially.
346 00:34:12.440 ⇒ 00:34:19.399 Uttam Kumaran: Yeah. And to talk about that kind of the way I’m thinking about. And this is what I told the operations team is is kind of similar is like.
347 00:34:19.780 ⇒ 00:34:22.350 Uttam Kumaran: In order to use AI effectively.
348 00:34:22.750 ⇒ 00:34:29.409 Uttam Kumaran: and like the most effective, we need to have a clear understanding of what the current process is
349 00:34:29.600 ⇒ 00:34:49.419 Uttam Kumaran: otherwise, it’s very hard to say, where can AI affect this in this example? Right? If if we have documented okay, every Pr gets made on Fridays, we go through, we look at the Pr. We then look at the lines that were changed. We compare them to the lines in the spreadsheet, and then we go update the fields.
350 00:34:49.620 ⇒ 00:34:55.540 Uttam Kumaran: If that’s the whole process, I will go. Take that process, hand it to the AI team, and say.
351 00:34:55.900 ⇒ 00:35:01.060 Uttam Kumaran: like, Make this cut this in time by into 10% of the time.
352 00:35:01.470 ⇒ 00:35:23.949 Uttam Kumaran: That will be my direction. Right? Like I will work. I’ll work on them. I’ll work with them on that. But that’ll sort of be how we start to automate a lot of these processes. I also think it’s still worth us doing it manually for at least a week or 2, because we will find out whether that I still not 100% sure that way we have the spreadsheet set up is
353 00:35:24.240 ⇒ 00:35:25.690 Uttam Kumaran: the right way.
354 00:35:27.570 ⇒ 00:35:33.029 Uttam Kumaran: Like. And it meaning like, are the columns there? All right? Should we have less stuff.
355 00:35:33.360 ⇒ 00:35:37.137 Uttam Kumaran: you know, like what’s important to have there?
356 00:35:38.650 ⇒ 00:35:43.659 Uttam Kumaran: because ultimately, like I even see the spreadsheet, because the spreadsheet is all static information
357 00:35:43.960 ⇒ 00:35:50.719 Uttam Kumaran: I like like, I guess the spreadsheet is is not static information, but it’s information we’re getting from somewhere, right like.
358 00:35:50.860 ⇒ 00:35:57.900 Uttam Kumaran: except for the dashboard questions. And like the dashboard metrics.
359 00:35:58.550 ⇒ 00:36:02.669 Uttam Kumaran: everything else we’re basically can pull from an Api
360 00:36:02.820 ⇒ 00:36:10.640 Uttam Kumaran: like we can pull the connections in 5 trainer portable from an Api, we can pull the code base. We can pull from Snowflake.
361 00:36:11.210 ⇒ 00:36:19.519 Uttam Kumaran: So I’m trying to think about. Okay, what is what is actually the most manual part of this spreadsheet? And what is the part that we have to get? Super right?
362 00:36:19.680 ⇒ 00:36:27.169 Uttam Kumaran: Seems like. It’s like what metrics are available, what dimensions are available, where to get them and what their definitions are.
363 00:36:27.340 ⇒ 00:36:34.210 Uttam Kumaran: And finally, the dashboard documentation. I don’t know if I have a solve for that yet.
364 00:36:34.600 ⇒ 00:36:37.320 Uttam Kumaran: But again, right now, we have, like
365 00:36:37.720 ⇒ 00:36:42.850 Uttam Kumaran: we have like 8 tabs. So that’s the thing. We think we should do it for 2 weeks, and see how it goes.
366 00:36:43.620 ⇒ 00:36:53.579 Caio Velasco: Okay, now we can, we can definitely start with that. And one thing that came up to my mind now is, for example, when you are using Meta base. I understand that for each.
367 00:36:53.730 ⇒ 00:36:57.220 Caio Velasco: Well, let’s call it metric. You’re you’re you’re creating.
368 00:36:57.798 ⇒ 00:37:06.540 Caio Velasco: It’s they understand. As a question in in Meta Base. So so I remember that when I was
369 00:37:06.830 ⇒ 00:37:10.349 Caio Velasco: looking at the questions on my tickets for Borges.
370 00:37:11.850 ⇒ 00:37:16.189 Caio Velasco: those are not necessarily obviously like a metric or something, but they’re just.
371 00:37:16.190 ⇒ 00:37:16.700 Uttam Kumaran: Correct.
372 00:37:16.700 ⇒ 00:37:20.070 Caio Velasco: Like guidance. Right and.
373 00:37:20.070 ⇒ 00:37:24.409 Uttam Kumaran: And this, and it’s also like, sorry not to cut you off. It’s it’s different in different bi tools.
374 00:37:24.830 ⇒ 00:37:25.799 Caio Velasco: Yeah, like, okay.
375 00:37:25.800 ⇒ 00:37:28.819 Uttam Kumaran: In tableau. It’s not organized like
376 00:37:28.960 ⇒ 00:37:37.109 Uttam Kumaran: it’s not technically organized as like a in metabase. They literally say, Question right, which is which is interesting. But
377 00:37:37.270 ⇒ 00:37:51.100 Uttam Kumaran: and the bi tools, it’s it’s just like you have metrics and dimensions. The question piece. I actually, it’s a way to force the analysts to think about the business question, because the way commonly analysts
378 00:37:51.410 ⇒ 00:37:58.690 Uttam Kumaran: go. And this also happens on the Ae side is like, okay, just write whatever SQL is asked. Don’t ask why
379 00:37:59.110 ⇒ 00:37:59.640 Uttam Kumaran: exactly.
380 00:37:59.640 ⇒ 00:38:08.979 Uttam Kumaran: I want to force it. So the why has to happen. And so if the spreadsheet is empty, meaning there’s no why we don’t want. We don’t work on it. So that’s the contract.
381 00:38:09.260 ⇒ 00:38:10.150 Uttam Kumaran: right?
382 00:38:10.370 ⇒ 00:38:13.619 Uttam Kumaran: And so that’s that’s sort of how I’m thinking, which is.
383 00:38:14.340 ⇒ 00:38:21.190 Uttam Kumaran: I think I want the I want the analysts to sort of work with the business to literally ask them, okay, what?
384 00:38:21.330 ⇒ 00:38:23.130 Uttam Kumaran: What questions do you have?
385 00:38:23.430 ⇒ 00:38:35.419 Uttam Kumaran: And then and then their process will evolve to where they’re saying, Okay, that’s a good question. Okay, that’s not really important question. Okay, here’s a really tough question. And then that gives us enough. That gives us a lot of
386 00:38:35.870 ⇒ 00:38:37.450 Uttam Kumaran: business understanding.
387 00:38:38.100 ⇒ 00:39:01.800 Uttam Kumaran: The toughest part about this business is, there’s so many layers right? There’s me project manager, there’s an analyst, there’s the Ae, there may be a data engineer. There’s the vendors, there’s the client. There’s so much loss right in this. In between the steps like playing the game telephone if you played that game. And so we want to make sure that everybody at every step has the ability to get all the context they need
388 00:39:02.298 ⇒ 00:39:05.680 Uttam Kumaran: and that’s why I love the spreadsheet, because it forces
389 00:39:05.850 ⇒ 00:39:08.749 Uttam Kumaran: us to do that. It forces the analysts to do that as well.
390 00:39:09.500 ⇒ 00:39:16.430 Caio Velasco: No, exactly putting everyone on board. I totally agree with that. Okay, okay. I’ll.
391 00:39:17.560 ⇒ 00:39:21.250 Uttam Kumaran: Well, I don’t know. I’ll I’ll have to start somewhere, and then that’s the work.
392 00:39:21.250 ⇒ 00:39:28.539 Uttam Kumaran: Start somewhere. It’ll be painful, I think. Start there. I think also, if you can, I know, I guess we were talking about time as well. So
393 00:39:29.240 ⇒ 00:39:38.400 Uttam Kumaran: right now, we’re basically we have. We have the scope of about 20 h for both
394 00:39:38.600 ⇒ 00:39:41.570 Uttam Kumaran: pool parts and Javi.
395 00:39:42.020 ⇒ 00:39:52.530 Uttam Kumaran: I think one thing that we should talk about between those 2 teams is sort of how the hours are getting split up. I never wanted I mean, previously we were really strict on like
396 00:39:52.690 ⇒ 00:39:58.399 Uttam Kumaran: you could only work because we weren’t. We wouldn’t make any money. Now, I’m sort of. There’s some flexibility.
397 00:39:58.580 ⇒ 00:40:12.219 Uttam Kumaran: but we can’t spend all of our hours on Javi just on documentation, right? There has to be other outputs. So I think that’s going to be the thing that I tasked the project manager to figure out for. Now don’t worry about it.
398 00:40:12.420 ⇒ 00:40:23.479 Uttam Kumaran: I would just do whatever you feel is right. I think it would help if you’re able to spend a few hours on both pool parts and Javi every day, at least for the next week.
399 00:40:23.930 ⇒ 00:40:24.570 Caio Velasco: Okay.
400 00:40:24.570 ⇒ 00:40:25.789 Uttam Kumaran: All in sort of
401 00:40:26.030 ⇒ 00:40:34.129 Uttam Kumaran: question, process, documentation world. I think you’re. I actually feel I’m kind of feel jealous that you’re able to do that. I wish I could do that because it’s like.
402 00:40:34.130 ⇒ 00:40:34.480 Caio Velasco: I just.
403 00:40:34.480 ⇒ 00:40:41.550 Uttam Kumaran: Calm place to go look, and there’s and like spend time poking around. But also I’m really hoping that
404 00:40:41.900 ⇒ 00:40:45.460 Uttam Kumaran: you can dive into that, and then come up with a lot of answers about
405 00:40:45.840 ⇒ 00:40:49.150 Uttam Kumaran: how we should set up documentation for every client.
406 00:40:49.380 ⇒ 00:40:55.290 Uttam Kumaran: And then those 2 clients are. We’re gonna probably after next week we’ll step on the gas again.
407 00:40:55.510 ⇒ 00:40:57.900 Uttam Kumaran: So I want to see how much progress we can make.
408 00:40:58.720 ⇒ 00:41:03.610 Caio Velasco: Okay, no great, great, great. I’ll think about some things, and I’ll start.
409 00:41:03.830 ⇒ 00:41:06.080 Caio Velasco: Put you on the Channel for doubts. The question.
410 00:41:06.420 ⇒ 00:41:23.910 Uttam Kumaran: Yeah, and just keep like, live blogging onto the channel about how what you’re finding. The other thing is also, if you can think about how to break up a task like that. If I’m like, okay, we wanna, we want to document this client. What are what are the steps? What are each of the artifacts like, okay, maybe we need a fig jam.
411 00:41:24.210 ⇒ 00:41:36.909 Uttam Kumaran: a visual like architecture. Diagram. Okay? Maybe we need the data platform documentation for every, for that whole client, everything filled out. Okay, we also need
412 00:41:37.700 ⇒ 00:41:41.300 Uttam Kumaran: something in notion, like what goes in notion. Is it per source?
413 00:41:41.460 ⇒ 00:41:42.349 Uttam Kumaran: Is it?
414 00:41:42.690 ⇒ 00:41:46.579 Uttam Kumaran: Is it like per business domain. And then, just like Faqs.
415 00:41:47.057 ⇒ 00:41:52.450 Uttam Kumaran: That would be helpful that way. Anyone, because there’s other clients where we’ll have to do this
416 00:41:52.580 ⇒ 00:42:03.320 Uttam Kumaran: like across Stack Blitz. We want to do this across Eden. We want to do this across our next client. And so I want to make sure those tasks are clear, because then we can. We can add those to.
417 00:42:03.910 ⇒ 00:42:05.160 Uttam Kumaran: you know, to the team.
418 00:42:05.900 ⇒ 00:42:07.280 Caio Velasco: Okay. Cool.
419 00:42:08.180 ⇒ 00:42:08.750 Luke Daque: So, just.
420 00:42:08.750 ⇒ 00:42:09.900 Uttam Kumaran: Keep that in the back of your mind.
421 00:42:10.340 ⇒ 00:42:17.616 Luke Daque: Yes, also for pool parts, since, like, there’s not much more data modeling going on. Maybe we can tackle.
422 00:42:18.540 ⇒ 00:42:21.830 Luke Daque: what do you call this, like some of the tech depths, or like.
423 00:42:22.310 ⇒ 00:42:22.860 Uttam Kumaran: Yes.
424 00:42:22.860 ⇒ 00:42:29.019 Luke Daque: Like, since, like the models were created before we had standardized how they should be. Maybe we can
425 00:42:30.305 ⇒ 00:42:33.590 Luke Daque: change the the yeah, the structure
426 00:42:33.700 ⇒ 00:42:36.640 Luke Daque: and stuff like that. Maybe that’s something we can do. Yeah.
427 00:42:38.260 ⇒ 00:42:48.098 Uttam Kumaran: Yeah, I totally agree. I think, as we start to apply our new documentation and our new naming conventions, we have legacy clients that need to get migrated.
428 00:42:49.280 ⇒ 00:42:55.779 Uttam Kumaran: so I would. Yeah, I would love to like if we see everything, then it’s then it’s clear to create the migration plan.
429 00:42:58.610 ⇒ 00:43:02.499 Caio Velasco: Yeah. My, my guess is that if we start
430 00:43:02.900 ⇒ 00:43:11.919 Caio Velasco: there’s asking questions regarding this documentation process, and people don’t know the answers right away. Then we are on the right path.
431 00:43:16.120 ⇒ 00:43:18.810 Caio Velasco: Because then we could definitely, actually a gap for that.
432 00:43:21.020 ⇒ 00:43:21.710 Luke Daque: Yeah.
433 00:43:25.780 ⇒ 00:43:32.969 Luke Daque: Core for stack with Utam. It’s also 20 h per week. Right?
434 00:43:36.340 ⇒ 00:43:42.930 Uttam Kumaran: for stock list. We are basically between 10 and 20. Right now I’m going to. We’re signing another contract with them.
435 00:43:44.060 ⇒ 00:43:44.610 Luke Daque: Okay.
436 00:43:45.000 ⇒ 00:43:47.010 Uttam Kumaran: So that’ll give us some more leeway.
437 00:43:47.650 ⇒ 00:43:48.420 Luke Daque: Gotcha.
438 00:43:50.000 ⇒ 00:43:52.109 Uttam Kumaran: Yeah, as you can see, there’s a lot of.
439 00:43:52.260 ⇒ 00:44:02.119 Uttam Kumaran: there’s a lot of stuff I need to do. So I appreciate you guys just holding the fort down, and that frees me up to go get these renewals, get the expansions, and get us some more.
440 00:44:02.410 ⇒ 00:44:07.439 Uttam Kumaran: Get us some more dollars to do this. I mean, one thing I’m really, really happy about this week is.
441 00:44:07.880 ⇒ 00:44:10.460 Uttam Kumaran: I feel really good about our team, like.
442 00:44:10.960 ⇒ 00:44:23.190 Uttam Kumaran: I think we have a mix of skill sets. We have a mix of like interests, not only on the engineering side, but also now I found you guys some really solid project managers.
443 00:44:23.621 ⇒ 00:44:29.149 Uttam Kumaran: That will make your when you wake up in the morning you will have a really clear understanding of like what needs to get done.
444 00:44:29.740 ⇒ 00:44:30.270 Luke Daque: Nice.
445 00:44:31.075 ⇒ 00:44:34.510 Uttam Kumaran: Which which is a real luxury like
446 00:44:34.790 ⇒ 00:44:43.379 Uttam Kumaran: I will say I don’t. I haven’t worked at many companies where that’s the case. In fact, I haven’t worked with any. I haven’t worked with a good project manager in my entire life. That’s why.
447 00:44:43.380 ⇒ 00:44:44.010 Caio Velasco: I became.
448 00:44:44.010 ⇒ 00:44:48.149 Uttam Kumaran: A product manager because I realized that
449 00:44:48.650 ⇒ 00:45:01.619 Uttam Kumaran: I’ve never worked with anyone good. And so I could just do the opposite of what everyone else did, and they would work. So we’ll see like that. May. That may end up being the case. I don’t know about you guys like, did you guys ever work with product or project managers before.
450 00:45:02.970 ⇒ 00:45:05.490 Luke Daque: No, that’s much as well for me.
451 00:45:05.490 ⇒ 00:45:10.949 Caio Velasco: Yeah, I I haven’t worked as a Pm. But I
452 00:45:11.900 ⇒ 00:45:16.579 Caio Velasco: there was a guy in in the the last consulting firm. I worked, and
453 00:45:16.710 ⇒ 00:45:19.199 Caio Velasco: he was the second one he was. He was good.
454 00:45:19.430 ⇒ 00:45:19.910 Caio Velasco: I like.
455 00:45:19.910 ⇒ 00:45:22.959 Uttam Kumaran: He’s good. Okay. What did he do? What did he do? Well.
456 00:45:23.260 ⇒ 00:45:27.139 Caio Velasco: So the 1st thing that he, when he started
457 00:45:27.340 ⇒ 00:45:43.170 Caio Velasco: it’s like, Hey, wait. So you’re building all this metrics and this dashboard. But you don’t have a document that defines the sign saying that we are building metric XY, and Z, and dashboard XY. And ZI mean, how how can we cover our our assets if something goes wrong on their side?
458 00:45:43.170 ⇒ 00:45:43.530 Caio Velasco: Yeah.
459 00:45:44.260 ⇒ 00:45:51.209 Caio Velasco: And I was like, Oh, my God, no one did that. True? So yeah, when he did, I was like, Okay, he, you know.
460 00:45:51.210 ⇒ 00:45:53.700 Uttam Kumaran: He an engineer like by background.
461 00:45:53.700 ⇒ 00:45:59.270 Caio Velasco: No, no, actually not is. I think of just a business background. Yeah.
462 00:46:00.500 ⇒ 00:46:03.319 Luke Daque: Are you still hiring for other project? Managers.
463 00:46:04.150 ⇒ 00:46:09.269 Uttam Kumaran: I’m happy to talk to any. I’m happy to talk to anybody. I don’t, I mean.
464 00:46:09.450 ⇒ 00:46:10.279 Uttam Kumaran: Fingers crossed.
465 00:46:10.280 ⇒ 00:46:12.670 Caio Velasco: America, but I think he lives in New York now.
466 00:46:13.270 ⇒ 00:46:21.709 Uttam Kumaran: Okay, yeah, I mean, look, I’m happy, like I, if there’s anyone in your network that you’re like, yeah, this person could be good or like
467 00:46:22.050 ⇒ 00:46:46.100 Uttam Kumaran: it. You don’t. You’re like not sure where they would fit. I’m happy to talk to them, whether we have a job for them right now or not. Not sure because I I hope that we don’t have to hire more people, because otherwise I’m gonna I’m gonna not be able to sleep anymore. When I look at when I look at the financial, so we made, we probably, I I think, between Steven and Amber.
468 00:46:46.430 ⇒ 00:46:57.089 Uttam Kumaran: we should be able to cover everything. But I want to talk to anybody you guys have, because we will continue to hire in the future. And I will. I want to build relationships with anyone that’s smart.
469 00:46:58.620 ⇒ 00:47:02.000 Uttam Kumaran: and sort of gets it, you know, that understands the trade-offs of
470 00:47:02.130 ⇒ 00:47:05.219 Uttam Kumaran: working with clients and and working in engineering and
471 00:47:05.510 ⇒ 00:47:19.450 Uttam Kumaran: sort of wants to work at a place like this. So if there’s anyone like that, yeah, you intro me at any time, like, even if we’re hiring or not, or whatever they do, it doesn’t. It’s not relevant to me. I just want to have a really good network of people like, you know that we have access to.
472 00:47:20.330 ⇒ 00:47:21.160 Caio Velasco: Great. Okay.
473 00:47:22.400 ⇒ 00:47:28.670 Uttam Kumaran: Like to give you an example. Akash, who you saw yesterday. Akash works at a company that’s like about 10
474 00:47:29.460 ⇒ 00:47:42.400 Uttam Kumaran: times bigger than us. But is a data consultancy. I wanted to hire him, but we don’t offer healthcare right now, and he’s like, I can’t join without that. And I’m like, okay, fuck. Well.
475 00:47:42.660 ⇒ 00:47:54.969 Uttam Kumaran: what can I like? I’m gonna we’re working like, I’m literally working on that right now. That is another thing on my to do list, which, as you guys probably know, I am not a healthcare expert. And so this is another thing I have to go learn. But
476 00:47:55.590 ⇒ 00:48:08.200 Uttam Kumaran: I was like, can you please? Just like, come on even part time, and just give us guidance on, like, how you guys run these engagements at your level. And so he’s like, Yeah, I’m happy to just come on for like 10 HA week or so. And this like.
477 00:48:08.900 ⇒ 00:48:18.089 Uttam Kumaran: so I basically am tasking him to go look at everything and find. Okay. Here’s actually like, for example, they probably have a the actual, like
478 00:48:18.520 ⇒ 00:48:22.170 Uttam Kumaran: scaled, tested version of this documentation problem.
479 00:48:22.570 ⇒ 00:48:35.590 Uttam Kumaran: And I’m kind of like, just give me the answer like, What is it? And so he’s coming on to help us. And then, ideally, when when we’re able to, you know. Figure out the benefits part, probably by the end of next quarter.
480 00:48:36.280 ⇒ 00:48:41.370 Uttam Kumaran: it’ll all be like I’ll bring them on. But that being said happy to talk to to anyone.
481 00:48:42.490 ⇒ 00:48:43.559 Caio Velasco: Okay. Cool.
482 00:48:44.080 ⇒ 00:48:44.680 Caio Velasco: Oh.
483 00:48:44.680 ⇒ 00:48:45.410 Uttam Kumaran: Okay, cool.
484 00:48:46.340 ⇒ 00:49:00.480 Uttam Kumaran: Okay. So I have to run to the next meeting. But yeah, just please, just like, keep as you guys doing everything keep talking in slack. I’m I’m actually very happy that I’m not the only one messaging in there. Of course, if you guys message me, I’m gonna say.
485 00:49:00.790 ⇒ 00:49:08.807 Uttam Kumaran: try messages in the group so that we keep some semblance of like like, we’re all in the same office together, you know. I think it’s really nice. And then, yeah, if
486 00:49:09.170 ⇒ 00:49:15.150 Uttam Kumaran: I think just as you have questions, just stream it there, and we’ll just keep discussing and and moving things along.
487 00:49:15.760 ⇒ 00:49:16.930 Caio Velasco: Perfect! Let me move!
488 00:49:17.500 ⇒ 00:49:18.130 Uttam Kumaran: Okay.
489 00:49:18.310 ⇒ 00:49:19.930 Uttam Kumaran: Alright, thanks. Guys. Have a good day.
490 00:49:19.930 ⇒ 00:49:21.390 Caio Velasco: Alright, thank you. Have a good day bye, bye.
491 00:49:21.390 ⇒ 00:49:22.070 Luke Daque: Bye-bye.
492 00:49:22.370 ⇒ 00:49:22.930 Uttam Kumaran: Bye.