Meeting Title: Uttam <> Ryan - Check-in Date: 2024-02-19 Meeting participants: Ryan Luke Daque, Uttam Kumaran
WEBVTT
1 00:00:36.760 ⇒ 00:00:39.220 Ryan Luke Daque: Hello, hey, hey?
2 00:00:39.750 ⇒ 00:00:41.650 Uttam Kumaran: How’s it going after colds?
3 00:00:41.760 ⇒ 00:00:47.369 Ryan Luke Daque: Huh? Yeah. Yeah. Sore throat. And like.
4 00:00:47.990 ⇒ 00:00:49.309 Uttam Kumaran: yeah, I don’t know.
5 00:00:49.920 ⇒ 00:00:52.930 Ryan Luke Daque: Yeah, we had. We had that last week as well, like
6 00:00:53.360 ⇒ 00:00:59.850 Ryan Luke Daque: everybody’s having sore throat. The I like the codes here because of the weather. It’s weird, but
7 00:01:00.400 ⇒ 00:01:06.509 Uttam Kumaran: yeah, it sucks. I just like I don’t know. I gotta go to the gym for the Sauna, like
8 00:01:06.680 ⇒ 00:01:09.930 Ryan Luke Daque: it’ll help
9 00:01:10.210 ⇒ 00:01:11.460 Uttam Kumaran: try to sleep.
10 00:01:12.400 ⇒ 00:01:13.310 Ryan Luke Daque: Nice.
11 00:01:13.710 ⇒ 00:01:23.509 Uttam Kumaran: Yeah, no. I just wanted to talk for a few minutes. Well, let me get a charge over. I just wanted to talk for a few minutes and see how to go, and we haven’t talked
12 00:01:23.630 ⇒ 00:01:37.300 Uttam Kumaran: like at length, until, since you know, we kind of started working more hours. So I just wanted to kind of get your perspective and see how things are going. I also have an idea, for
13 00:01:37.540 ⇒ 00:01:46.430 Uttam Kumaran: I have an opportunity with another client that I wanted to actually ask you about kind of wanted to gauge your bandwidth, and then
14 00:01:46.590 ⇒ 00:01:48.340 just get a sense of like.
15 00:01:48.550 ⇒ 00:01:57.040 Uttam Kumaran: what are the what are the tasks that have been working well with with pool parts. Then, yeah, just get a sense of how everything’s going.
16 00:01:57.530 ⇒ 00:01:58.489 Ryan Luke Daque: Yes, sir.
17 00:02:00.210 ⇒ 00:02:05.920 Ryan Luke Daque: yeah. I think it’s it’s been going well, so far, like, yeah, I’ve been. I’ve been adapting like the first.
18 00:02:06.210 ⇒ 00:02:15.870 Ryan Luke Daque: I guess. Well, I I’m still like in the like the change curve, I guess, in terms of like trying to adapt with it within the number of hours. Especially.
19 00:02:16.020 ⇒ 00:02:23.200 Ryan Luke Daque: II also started going to the gym, so that like, yeah, like something like that, and
20 00:02:23.260 ⇒ 00:02:32.820 Ryan Luke Daque: like like the first week, I believe, went well, at at least for February, like we’re we’re doing like data models and stuff like that the last week was pretty
21 00:02:33.070 ⇒ 00:02:38.099 Ryan Luke Daque: like new to me, like in terms of like trying to figure out stuff. And
22 00:02:38.260 ⇒ 00:02:44.990 Ryan Luke Daque: I think I’ve been like over thinking most of the time, and and that like also cost to be
23 00:02:45.040 ⇒ 00:02:56.420 Ryan Luke Daque: like my output to be a lot slower or like, Yeah, III was like just over thinking I was like creating queries out of nothing and like not getting their results and stuff like that. So.
24 00:02:56.550 ⇒ 00:03:10.240 Ryan Luke Daque: yeah, that was pretty slow. But I think II got I figured it out like towards the end of the week. And and yeah, I think we should be good to go so far. But yeah, in terms of the hours, I think. It’s good.
25 00:03:10.540 ⇒ 00:03:20.070 Ryan Luke Daque: Yeah. The more the more time I like spend looking at the data, the sources, the tables that we have, the the more I like
26 00:03:20.210 ⇒ 00:03:26.030 Ryan Luke Daque: kind of understand, like the the you know, like
27 00:03:27.140 ⇒ 00:03:37.610 Ryan Luke Daque: probably the best way to not not do the best way, but like, how? How they like connect to each other, basically something like that. So yeah.
28 00:03:37.850 ⇒ 00:03:46.089 Uttam Kumaran: how do you feel about all the different like types of work that we’ve done right. We’ve done like some stuff on the Etl side. We’ve done
29 00:03:46.120 ⇒ 00:03:51.819 Uttam Kumaran: like testing. We’ve done actual modeling. And now some analysis like.
30 00:03:51.840 ⇒ 00:03:54.070 Uttam Kumaran: out of all that, like what’s been
31 00:03:54.230 ⇒ 00:04:01.430 Uttam Kumaran: what’s been the most like rewarding stuff, and like, what? Out of all those like 5, like, what are the things that you wanna continue to do?
32 00:04:03.190 ⇒ 00:04:04.040 Ryan Luke Daque: Hmm.
33 00:04:06.670 ⇒ 00:04:10.150 Ryan Luke Daque: like, yeah, all of them are like, pretty
34 00:04:11.460 ⇒ 00:04:22.360 Ryan Luke Daque: like, I basically like working on all of them. II think data modeling is basically the one thing I really love doing, like times feel so fast when I
35 00:04:22.490 ⇒ 00:04:29.249 Ryan Luke Daque: doing data modeling and like it’s it’s more rewarding to me to be able to get the results that are
36 00:04:29.560 ⇒ 00:04:41.120 Ryan Luke Daque: needed. Like doing tests as well, so that we make sure that there is the the data quality is good even the investigations. When we were doing like trying to check that
37 00:04:41.330 ⇒ 00:04:45.989 discrepancies between, like the Google sheets. And and what we have
38 00:04:46.210 ⇒ 00:04:55.769 Ryan Luke Daque: in the data model. That’s also like pretty rewarding, although it’s like. it’s consuming. I mean, that time consuming and like it makes you like
39 00:04:56.020 ⇒ 00:04:57.199 Ryan Luke Daque: speak a lot
40 00:04:57.500 ⇒ 00:05:01.239 Ryan Luke Daque: that. I guess the least that I
41 00:05:01.510 ⇒ 00:05:03.039 Ryan Luke Daque: like doing. I guess
42 00:05:03.290 ⇒ 00:05:15.029 Ryan Luke Daque: last week, because maybe that was pretty new to me. So I I’m not very used to doing doing that stuff. But but yeah, III can. I can see the value
43 00:05:15.120 ⇒ 00:05:25.060 Ryan Luke Daque: of what we’re doing there. Especially. Yeah, it’s just it’s just new to me. So yeah, it’s it’s a pretty great experience as well to be a to be
44 00:05:25.160 ⇒ 00:05:27.019 in the perspective
45 00:05:27.190 ⇒ 00:05:28.240 Ryan Luke Daque: of like.
46 00:05:29.350 ⇒ 00:05:36.540 Ryan Luke Daque: it’s basically like the client’s perspective, right? Right? Because they they want to. You want to know us
47 00:05:36.960 ⇒ 00:05:52.870 Ryan Luke Daque: stuff from the data that you can make business decisions on and stuff like that. So like, it’s, it’s like a higher, more up upstream work that’s being done before even data modeling before creating pipelines and stuff like that. So yeah.
48 00:05:52.980 ⇒ 00:06:01.670 Ryan Luke Daque: must be the interesting. And maybe so. after that, it’s like, maybe of visualization.
49 00:06:01.940 ⇒ 00:06:07.350 Ryan Luke Daque: Just and just because light dash isn’t very
50 00:06:07.990 ⇒ 00:06:20.030 Ryan Luke Daque: there’s not a lot of customization you can do in light dash. Basically, it’s like pretty fixed in in terms of what you can do. So you can’t really do much. But yeah, other than that, everything’s
51 00:06:20.160 ⇒ 00:06:21.440 Ryan Luke Daque: continuing. Well.
52 00:06:21.520 ⇒ 00:06:25.520 Uttam Kumaran: okay, good. No. It’s just good for me for my contacts. It’s just good to hear
53 00:06:25.580 ⇒ 00:06:41.409 Uttam Kumaran: what are the things that you feel like your strongest, that, and then also what are the things that interest you. And then for me, it’s understanding. Okay, cool out of the range of things that the company has like. Where can I make sure you’re plugged into the points where you’re doing your best work.
54 00:06:41.410 ⇒ 00:07:03.659 Uttam Kumaran: and then also, just make sure that like at any point, if you’re struggling to make sure that. Okay, how can I make sure to give you the resources, or, like, you know, help needed to kind of push through. And II would say, I agree. II could definitely notice that your skill set is is growing a lot in the data modeling. And I think even from when we started working towards
55 00:07:03.730 ⇒ 00:07:28.300 Uttam Kumaran: even like, you know, last month I could see tremendous differences, because I think a lot of the things that you learned while we were working together about like how to like, identify specific issues or all the work we’ve done on Github Prs, and like kind of the pushing process. I think, has been like, really, really good. And so I definitely see moves there. I know last week we’re kind of transitioning to more of this like analyst work.
56 00:07:28.500 ⇒ 00:07:57.320 Uttam Kumaran: And I was definitely, gonna I was definitely interested to see how you kind of handled it, because it is a little bit of a different. It’s like very open ended, which ask one question, then you ask another question, then you ask another question, and then it requires some bouncing back between like, okay, let me create dimension. Ask another question like, it’s a little bit free, flowing compared to the data modeling where there’s a really fixed end goal. And you can kind of see and that’s okay, like I would say.
57 00:07:57.400 ⇒ 00:08:00.359 Uttam Kumaran: out of all those things right? Like you start from
58 00:08:00.400 ⇒ 00:08:16.820 Uttam Kumaran: Etl to like modeling and doing stuff in the warehouse to Bi to like doing, testing and then doing analysis, like, not many people can do the whole thing right? So that’s something that, like I. It’s a it’s a lot to ask
59 00:08:16.820 ⇒ 00:08:33.439 Uttam Kumaran: for one person to kind of bounce through all that. So it’s I’d I’d for me. It’s just important to hear like, Hey, this is the stuff that I really like. However, I think it’s it’s good for me to kind of get that context to be able to know. Okay, cool. Ryan really plugs in here, really? Well, and then let me see.
60 00:08:33.650 ⇒ 00:08:38.650 Uttam Kumaran: for example, the reason why I went and found someone like Pat is because II would say
61 00:08:38.720 ⇒ 00:09:03.939 Uttam Kumaran: me or you. I don’t really like doing dashboarding that much, either. I’m okay at it, but it’s not like what I really love to do, and II will do it just to get the job done. But I wanted to bring on Pat, who’s like an expert in like dashboarding to be able to take that off our plate very similar. Yeah, that’s what he really loves doing. And I want him to kind of go and run freely and do that
62 00:09:04.000 ⇒ 00:09:31.080 Uttam Kumaran: very similarly, on the analyst side, like, I would say, I am okay, like I would say, my I. My love, too, is doing data, modeling and doing stuff in the warehouse and doing kind of like this devops stuff. But I’ve also done a lot of analyst work, and so I can go do that but again, my job at the moment is to find opportunities to like, bring a team together and then get me doing sales like I have about like 3 different sales calls today.
63 00:09:31.080 ⇒ 00:09:40.450 Uttam Kumaran: So I’m actually continuing to interview some people for like handling this analyst part, because I think again across me and you, I think we’re we’re okay at it.
64 00:09:40.450 ⇒ 00:10:09.550 Uttam Kumaran: But I wanna get someone who’s like that’s their number. One mission is like finding that you could do that work in like 10 times faster than us. So so one thing, I think it’s really good that just to get experience doing all those things that way. In case you need, we need to do something. We can all produce anything in that whole pie. We can all produce right looking for people who are their only skill set is like running sequel queries. Everybody should have understanding of how the entire cake is made. But
65 00:10:09.640 ⇒ 00:10:12.380 Uttam Kumaran: knowing which part you’re strong is that
66 00:10:12.400 ⇒ 00:10:19.149 Uttam Kumaran: is important, like, just because you can do it doesn’t mean you should do it right. So for me that I know. That’s like
67 00:10:19.240 ⇒ 00:10:34.179 Uttam Kumaran: again, when you run a team, you need people with specific roles and so kind of thing I wanted to ask. Today, I think still, this week. It may be a little bit of a transition week, because I’m just now kind of looking for some people that can come handle that analyst work. But
68 00:10:34.460 ⇒ 00:10:54.180 Uttam Kumaran: my question was, gonna be, I have another client that I need some data modeling work for. They have a web platform so that we’re getting like web events, from from their platform as well as some like product data. The the client is called Asset link.
69 00:10:54.220 ⇒ 00:10:58.460 Uttam Kumaran: It’s like a Ss ETLI, NK. AI,
70 00:10:58.730 ⇒ 00:11:05.140 Uttam Kumaran: It’s a web platform to connect financial advisors and wealth managers together.
71 00:11:05.400 ⇒ 00:11:15.620 Uttam Kumaran: and they need some data modeling work. I’ve been doing a lot of the data modeling work so far few months now, but it’s becoming a little bit
72 00:11:16.150 ⇒ 00:11:41.709 Uttam Kumaran: much for me to handle so I wanted to see whether we could split some of your time. Now that we’ve gone through a lot of the major data modeling tasks for pool parts. I think it’s gonna be a lot. It’s not gonna require a full like 8 HA day there. So I wanna see whether we can split your time between the 2 clients. And in a week like this, where, for example, it’s a lot of analysts work, and maybe I can sub in.
73 00:11:41.760 ⇒ 00:11:51.209 Uttam Kumaran: There’s a lot of work to be done for asset link. So wondering like if you’d be open to that, I know it’s a it’s a little bit of jumping between 2. But
74 00:11:51.480 ⇒ 00:11:54.140 Uttam Kumaran: again, I everything else kind of stays the same.
75 00:11:54.180 ⇒ 00:11:59.239 Uttam Kumaran: Yeah, I’d be down to that actually a lot.
76 00:11:59.320 ⇒ 00:12:02.099 Uttam Kumaran: It’s actually a lot easier. There’s only 2 sources.
77 00:12:02.110 ⇒ 00:12:07.949 Uttam Kumaran: And it’s everything else is the same like github structure. dB, structure, everything.
78 00:12:08.140 ⇒ 00:12:17.939 Uttam Kumaran: But it’s just a lot of it’s it’s just a lot of data modeling work. And then really basic light dash work. Kind of similar to some of the stuff we did in the very beginning for pull parts
79 00:12:18.060 ⇒ 00:12:21.790 Ryan Luke Daque: right? Yeah. Sure. I I’d be down to that.
80 00:12:24.690 ⇒ 00:12:28.250 Uttam Kumaran: So let me kind of give you a bit of an
81 00:12:28.680 ⇒ 00:12:35.560 Uttam Kumaran: Let me kind of give you access to everything. Maybe I can just show you right now, if that’s okay, I’ll just kind of show you the repo structure.
82 00:12:35.590 ⇒ 00:12:36.630 Ryan Luke Daque: That’s correct
83 00:12:37.740 ⇒ 00:12:40.000 Uttam Kumaran: show you what you have.
84 00:12:41.730 ⇒ 00:12:46.840 Uttam Kumaran: yeah. And then again, I’m interviewing 2 people today for kind of like analyst type work.
85 00:12:46.920 ⇒ 00:12:57.279 Uttam Kumaran: Basically, I was like, I wanted to share with them the questions we’re trying to answer and be like, How long do you think this would take you for me? I’m very. I’m like very direct, like
86 00:12:58.060 ⇒ 00:13:21.690 Uttam Kumaran: I. If you know how to do this stuff, you know how to do it. If you don’t know, people would be like, well, I don’t know. It’s like, so for me, I want to talk to a really good analyst like, if I was to ask you like to show you a data model problem be like, how long you could be like, it’s gonna take this long, similar side, as I ask the same thing to pat. And like I got, we go. So I wanna just get people who are like really experts. So I’m gonna interviewing 2 people.
87 00:13:21.690 ⇒ 00:13:30.250 Uttam Kumaran: I, you know, I look for some people, these both those one guys from Argentina, another guy is from he’s from the Us.
88 00:13:30.540 ⇒ 00:13:34.499 Uttam Kumaran: I don’t know. I didn’t have much luck finding anybody else really like
89 00:13:34.710 ⇒ 00:13:50.050 Uttam Kumaran: people in the Philippines or in India. Through my connections, it seems like people have like have like somewhat analysts, experience. But it was like an excel. But it wasn’t anything like advanced, you know, right?
90 00:13:50.710 ⇒ 00:13:55.909 Ryan Luke Daque: Like, it’s not easy to find. People who can do that. Actually.
91 00:13:56.100 ⇒ 00:13:59.750 Uttam Kumaran: yeah. And and another idea like, I wonder do you think like
92 00:13:59.770 ⇒ 00:14:17.409 Uttam Kumaran: cause? II know I asked you in the beginning, like for intros to people that you know, that may have similar skill set, but I wonder, like how I his, for example, I know a lot of. I have some other smart connections in Philippines and in India. But they just don’t know Dvt and data model. I almost wanna like, get them trained up
93 00:14:17.910 ⇒ 00:14:25.449 Uttam Kumaran: stuff. But like, I don’t know what the process is like, or if that’s something you think we should try to do? Because.
94 00:14:26.420 ⇒ 00:14:36.570 Uttam Kumaran: like, yeah, I mean, because again, I have some people here in the Us that could do that sort of work. But it’s very, very pricey right?
95 00:14:36.790 ⇒ 00:14:45.599 Uttam Kumaran: And so, and they’re not as hungry for the opportunity. And so, if I could if I could give that opportunity someone who’s smart somewhere else. I’m happy to
96 00:14:45.980 ⇒ 00:14:49.840 Uttam Kumaran: so I don’t know. Maybe that’s something we could think about longer term as well.
97 00:14:50.300 ⇒ 00:14:54.220 Ryan Luke Daque: Yeah, I can. Yeah, I’ll also try to find
98 00:14:54.310 ⇒ 00:15:06.869 Ryan Luke Daque: within my connections, but, like most of my connections, are also already have worked. So I’m not sure if they would like. Yeah. And again, I’m happy to meet with anyone, just to, you know.
99 00:15:06.940 ⇒ 00:15:09.459 Uttam Kumaran: Have a friendly conversation, and like.
100 00:15:09.720 ⇒ 00:15:13.519 Uttam Kumaran: say, whatever you do get free at some point, let me know. But
101 00:15:13.900 ⇒ 00:15:22.880 Uttam Kumaran: I don’t know for me. I just want to meet people who have done Dbt and Snowflake work anywhere in the world, wherever they are, and give the opportunity, because
102 00:15:23.540 ⇒ 00:15:44.669 Uttam Kumaran: there’s there’s a lot of people in the Us. That can do it. But they don’t. They’re not like, I kinda wanna just open it up to whoever we can get and kind of building relationships with people that know this sort of work. So you know, I actually what I did is the other day I went on to the Dvt slack and I went to jobs. And then I just saw people who were saying, Hey, I’m open to work.
103 00:15:44.870 ⇒ 00:15:50.610 Uttam Kumaran: and and I just go look at their github, and I kind of figure out if they’re good. And then I just like message call people.
104 00:15:50.730 ⇒ 00:15:56.410 Uttam Kumaran: So like, there’s a bunch of ways we can find these folks. And there’s like meetups and stuff, too. So
105 00:15:57.900 ⇒ 00:16:00.130 Ryan Luke Daque: yeah, that’s a good idea.
106 00:16:02.100 ⇒ 00:16:04.780 Okay, so let me share.
107 00:16:05.970 ⇒ 00:16:06.890 Ryan Luke Daque: Let’s see.
108 00:16:10.040 ⇒ 00:16:14.269 Ryan Luke Daque: this is ass asset link AI, right?
109 00:16:14.290 ⇒ 00:16:16.130 Uttam Kumaran: Yeah, this is asset link
110 00:16:17.320 ⇒ 00:16:21.800 Ryan Luke Daque:
111 00:16:22.710 ⇒ 00:16:28.739 so they’re previously called Dfd, they just change their name. So you will see Dfd or Acid link everywhere.
112 00:16:28.960 ⇒ 00:16:36.039 Uttam Kumaran: But basically, there’s there’s a Dbt project. So there’s a couple of things.
113 00:16:36.930 ⇒ 00:16:46.640 Uttam Kumaran: The main thing that I worked on for them for a couple months was. they have this platform asset link. I’ll I’ll log in and show you the product. And then you could also create an account.
114 00:16:50.460 ⇒ 00:17:00.589 Uttam Kumaran: So they have advisors and asset and asset managers. So wealth, managers, asset managers, and you could come in and you create a profile, and the goal is for you to match up
115 00:17:00.600 ⇒ 00:17:20.240 Uttam Kumaran: you. You’re someone that that managers money, and there you are, someone who invest the money, and you want to be able to match those people up together, and so kind of built the platform to do so. Here’s like my account here on the right. That’s Devon. He’s the CEO and you create a profile. And then up here you can also search for profiles.
116 00:17:20.470 ⇒ 00:17:23.910 Uttam Kumaran: you know, this is another.
117 00:17:23.930 ⇒ 00:17:33.629 Uttam Kumaran: a friend of mine who actually works on. Who’s the CTO, that company? So basically, what I worked on for a few months is actually this score.
118 00:17:35.110 ⇒ 00:17:42.679 Uttam Kumaran: So these scores that you see here are actually coming from Snowflake and coming from a Dbt job. It’s coming from
119 00:17:42.940 ⇒ 00:17:55.870 Uttam Kumaran: this affinity score called results production. What this is doing is it’s taking all of the user data. So I’m getting it all the user data for wealth managers and asset managers.
120 00:17:56.740 ⇒ 00:17:59.189 Uttam Kumaran: I then wrote like an algorithm to
121 00:17:59.350 ⇒ 00:18:04.259 Uttam Kumaran: create an affinity score between based on the answers to their profiles.
122 00:18:04.420 ⇒ 00:18:08.440 Uttam Kumaran: So, for example, if there’s a match between, they’re both us veterans.
123 00:18:08.550 ⇒ 00:18:18.420 Uttam Kumaran: They’re both like they have certain like faith based investment. Like, you know, preferences.
124 00:18:18.500 ⇒ 00:18:20.339 Uttam Kumaran: We create a map score
125 00:18:20.380 ⇒ 00:18:40.469 Uttam Kumaran: and then this is like a really interesting script. This took like a couple of months to kind of like, go through and test and develop, and then I’m able to create like a score that matches them. So, for example, if I go here and I could see like me, and surf match 98. If I look in the database, you’ll not only see the 98, but you’ll see why we matched
126 00:18:40.810 ⇒ 00:18:47.380 Uttam Kumaran: like to keep like which which categories you match and stuff like that? Yeah, exactly. So. This is a script that we wrote.
127 00:18:47.490 ⇒ 00:19:02.129 Uttam Kumaran: The other scripts that’s coming in is that like this ask manager users. So this is just bringing in. So from the product table, from the from the product data, we get users table. So this is like all the users that are on the platform.
128 00:19:02.150 ⇒ 00:19:13.230 Uttam Kumaran: Right? People like serve people like me. And there’s 2 class of users. There’s the asset managers and the wealth managers. And so basically involved the profile data. So you can see, like
129 00:19:13.460 ⇒ 00:19:18.180 Uttam Kumaran: all this different data. And then I’m filtering to just asset managers.
130 00:19:18.260 ⇒ 00:19:24.370 Uttam Kumaran: Similar table here called wealth manager users, where I’m just bringing in all the data
131 00:19:24.450 ⇒ 00:19:29.570 Uttam Kumaran: filtering to wealth managers the other source of data is the web events.
132 00:19:30.270 ⇒ 00:19:31.370 Uttam Kumaran: So
133 00:19:31.420 ⇒ 00:19:36.869 Uttam Kumaran: they have a web events platform that’s bringing in. I don’t know if you ever worked with web events before, but it’s like
134 00:19:36.910 ⇒ 00:19:47.039 Uttam Kumaran: page views. host the device what page they’re on. So I’m just bringing in all those events here. And then, if we go to
135 00:19:47.350 ⇒ 00:19:53.729 Uttam Kumaran: light dash, I’ll show you kind of like the state of light. Dash right now.
136 00:20:04.830 ⇒ 00:20:09.100 Uttam Kumaran: Great. So this is like,
137 00:20:10.580 ⇒ 00:20:14.790 Uttam Kumaran: So this is like, currently our our current like web platform dashboard.
138 00:20:15.280 ⇒ 00:20:16.500 Uttam Kumaran: It’s just showing
139 00:20:17.190 ⇒ 00:20:19.110 Uttam Kumaran: where people are coming from.
140 00:20:19.360 ⇒ 00:20:21.630 Uttam Kumaran: what pages they’re visiting.
141 00:20:22.660 ⇒ 00:20:26.349 Uttam Kumaran: where, like, where they’re where they’re using the product from things like that.
142 00:20:28.500 ⇒ 00:20:35.930 Uttam Kumaran: So all based on web events. And there’s also a table here for the actual affinity scores.
143 00:20:36.870 ⇒ 00:20:44.549 Uttam Kumaran: So, for example, if I put in the asset manager, name, wealth, manager name and the affinity score. You’re gonna see
144 00:20:46.840 ⇒ 00:20:55.859 Uttam Kumaran: all the different matches. So if you think about it. the amount of matches is actually a multiplication, right. So if you have a thousand asset managers.
145 00:20:55.910 ⇒ 00:20:59.619 Uttam Kumaran: they need a score between all 1,000 and all the other
146 00:20:59.630 ⇒ 00:21:02.629 Uttam Kumaran: wealth managers. So you can see there’s like
147 00:21:02.830 ⇒ 00:21:07.960 Uttam Kumaran: there’s a ton of different affinity scores. But, for example, if I take one
148 00:21:08.280 ⇒ 00:21:15.819 Uttam Kumaran: asset manager, you’ll see all his matches between him and all the rest. Yeah. And it’s going to be yeah
149 00:21:16.220 ⇒ 00:21:21.540 Ryan Luke Daque: nice. And then the one thing I brought in is like a match reasoning so you could see why they matched
150 00:21:21.900 ⇒ 00:21:23.200 Ryan Luke Daque: gotcha
151 00:21:24.420 ⇒ 00:21:29.280 Uttam Kumaran: and then what? So what’s happening is, I’m bringing the data through 5 tram
152 00:21:29.420 ⇒ 00:21:35.070 Uttam Kumaran: I’m running my match. And then I actually send surf an file.
153 00:21:35.630 ⇒ 00:21:36.580 Uttam Kumaran: And I’m good
154 00:21:36.680 ⇒ 00:21:44.290 Uttam Kumaran: using Snowflake copy. So if you I don’t know if I just left the Github. But if I go to Github
155 00:21:47.370 ⇒ 00:21:52.750 Uttam Kumaran: I wanted to show you where that just so, you have an understanding of kind of the data pipeline.
156 00:21:55.800 ⇒ 00:21:59.460 Uttam Kumaran: So at the top of this.
157 00:21:59.930 ⇒ 00:22:05.899 Uttam Kumaran: Dfd, affinity results production, there’s a post hook. Have you used a post?
158 00:22:06.480 ⇒ 00:22:08.849 Ryan Luke Daque: I haven’t. But yeah, I
159 00:22:09.060 ⇒ 00:22:14.620 Ryan Luke Daque: kind of understand what. Basically, it’s just like, run this right after this, this finishes
160 00:22:14.770 ⇒ 00:22:16.330 Ryan Luke Daque: right.
161 00:22:16.370 ⇒ 00:22:19.930 Uttam Kumaran: I have like right after this finishes it, takes
162 00:22:20.030 ⇒ 00:22:24.470 Uttam Kumaran: the file and puts it into this S. 3 bucket
163 00:22:24.510 ⇒ 00:22:26.770 Ryan Luke Daque: right as a Csv.
164 00:22:27.210 ⇒ 00:22:30.590 Uttam Kumaran: and then the back end team will go and grab this file.
165 00:22:30.990 ⇒ 00:22:34.039 Ryan Luke Daque: Gotcha, and where’s the S. 3 bucket stored?
166 00:22:34.210 ⇒ 00:22:36.470 Uttam Kumaran: It’s an aws
167 00:22:36.820 ⇒ 00:22:40.810 Uttam Kumaran: I have it set up with them through my account. But
168 00:22:41.880 ⇒ 00:22:54.850 Uttam Kumaran: yeah, so so surf surf is the CTO of that company. He worked together, actually at a previous company, and so he he kind of runs all the back end and design team
169 00:22:55.110 ⇒ 00:22:55.910 Uttam Kumaran: so
170 00:22:57.160 ⇒ 00:23:09.830 Ryan Luke Daque: nice and like, W. What’s the source? Do you have like? Have access like you have the the page views and stuff like that. That’s all in snowflake. So I’ll give you access to
171 00:23:11.040 ⇒ 00:23:12.150 Uttam Kumaran: I can show you.
172 00:23:21.890 ⇒ 00:23:24.870 Uttam Kumaran: Let’s see, maybe I even have a Yes code.
173 00:23:45.730 ⇒ 00:23:55.480 Ryan Luke Daque: but, like the actual source is just, is also like coming from them, and they just send it to us in Snowflake, or something like you have a.
174 00:23:56.260 ⇒ 00:23:59.809 Uttam Kumaran: So there’s 2 sources. There’s 5 trans for the product data.
175 00:23:59.890 ⇒ 00:24:07.210 Uttam Kumaran: So you’ll see here, there’s an aurora postgres. This is the postgres database for the product
176 00:24:07.370 ⇒ 00:24:09.130 Ryan Luke Daque: right?
177 00:24:10.100 ⇒ 00:24:12.409 and I don’t know whether this is gonna load.
178 00:24:18.630 ⇒ 00:24:23.160 Ryan Luke Daque: Yeah. So you see the 2 tables. So there’s profile and user table
179 00:24:23.290 ⇒ 00:24:33.580 Uttam Kumaran: and then post hog, post hog is the web events platform. They’re using a direct link into Snowflake, and they output an event table here.
180 00:24:33.760 ⇒ 00:24:34.690 Ryan Luke Daque: Nice?
181 00:24:35.160 ⇒ 00:24:44.189 Uttam Kumaran: yeah. And then I also have
182 00:24:47.920 ⇒ 00:24:50.080 Uttam Kumaran: a very similar
183 00:24:54.080 ⇒ 00:24:55.000 Uttam Kumaran: see?
184 00:25:01.660 ⇒ 00:25:09.930 Uttam Kumaran: So
185 00:25:12.840 ⇒ 00:25:15.780 Uttam Kumaran: within light dash to the other tables that are there.
186 00:25:16.710 ⇒ 00:25:20.170 You’ll see are the asset managers, wealth managers, and the web events
187 00:25:20.330 ⇒ 00:25:28.659 Uttam Kumaran: web events is a ton of data. So I probably need some help. I’ll I’ll kind of show you the tickets we have set up, and we can kind of run through the kind of a very similar process. But.
188 00:25:29.080 ⇒ 00:25:35.840 Uttam Kumaran: we have a ton of web events that are being brought in. This will require a little bit of modeling to clean up.
189 00:25:36.090 ⇒ 00:25:39.669 Ryan Luke Daque: I just bought everything in recently in the last few weeks.
190 00:25:40.310 ⇒ 00:25:48.660 Uttam Kumaran: And then, apart from that, everything else is running pretty much the standard way.
191 00:25:48.860 ⇒ 00:25:56.860 Uttam Kumaran: I have a project for them that I kind of ever been using for my work.
192 00:25:57.190 ⇒ 00:26:02.189 Uttam Kumaran: but we can start to use this in a very similar way.
193 00:26:02.380 ⇒ 00:26:03.260 Ryan Luke Daque: Okay.
194 00:26:03.450 ⇒ 00:26:13.190 Uttam Kumaran: So let me get you access to everything, and then we can kinda maybe talk later today, or maybe tomorrow, about just like a couple of tasks here.
195 00:26:13.300 ⇒ 00:26:20.689 Uttam Kumaran: I think that’s kind of it. I’ll add you to the channel and things like that.
196 00:26:21.720 ⇒ 00:26:36.250 Uttam Kumaran: yeah. So the I would say, the major tasks here are like, I wanna do some modeling for the web events so like it’s like, gonna require some cleanup, a couple of new metrics. I also want to begin to link the web events to the profiles.
197 00:26:36.740 ⇒ 00:26:46.210 Uttam Kumaran: So, for example, when a user comes on and they’re logged in and they click on something, I will tie that web event back to that user
198 00:26:46.330 ⇒ 00:26:54.980 Ryan Luke Daque: some work either from our side or from their side, or figuring out how they can provide us as part of the event properties.
199 00:26:55.620 ⇒ 00:27:12.679 Uttam Kumaran: The user id of the person that triggered the event. For example, I’m login to Github right now. If I click on this on their end, they’re gonna be able to sell. Tell that I’m the one that clicked on that action. Yeah, I want to be able to do that as well.
200 00:27:13.080 ⇒ 00:27:21.740 Uttam Kumaran: And then the last thing. So we’re working on almost like 3 different, like core dashboards for these guys, one is like overview of the web events.
201 00:27:21.810 ⇒ 00:27:29.090 Uttam Kumaran: Second is an overview of the users on platform. So if I go to asset link users
202 00:27:29.280 ⇒ 00:27:38.590 Uttam Kumaran: right now, I just have a couple of tiles that show how many asset managers. There are the firms that they work at
203 00:27:38.670 ⇒ 00:27:46.569 Uttam Kumaran: and the new ones that are creating profile. So there’s 90 currently here are all 90. And then here’s the companies that they all work for.
204 00:27:48.290 ⇒ 00:27:59.730 Uttam Kumaran: So I’m this is another dashboard that we’re working on is just who are the users on platform. How many users are there? How many are wealth managers? Ask the managers who do they work for? And then how is that growing over time?
205 00:27:59.900 ⇒ 00:28:13.000 Uttam Kumaran: The third dashboard that we’re working for working on for them is a customer facing dashboard. So one thing is, they want to actually sell a dashboard to their clients
206 00:28:13.110 ⇒ 00:28:25.549 Uttam Kumaran: on the platform that show. hey? For on your profile you have this many views, this many messages. Here’s like, how filled out your profile is so like, what would a customer see about like their usage on the platform.
207 00:28:25.930 ⇒ 00:28:32.710 Uttam Kumaran: right? So that’s the third dashboard that we’re kind of working on. So those are the 3 dashboards that I’m working with them on
208 00:28:32.930 ⇒ 00:28:38.410 Uttam Kumaran: and so we’ll kind of create tickets around, you know all those 3. So
209 00:28:38.990 ⇒ 00:28:44.559 Uttam Kumaran: so let me see how how quickly I can get you access to everything today.
210 00:28:44.780 ⇒ 00:28:45.730 Ryan Luke Daque: thank you.
211 00:28:46.230 ⇒ 00:28:51.880 Uttam Kumaran: Yeah, we can just think about splitting time between pool parts.
212 00:28:51.940 ⇒ 00:29:13.290 Uttam Kumaran: And this is, I would say on the pool part side, I’m gonna I’ll kind of take on. Probably some of the analysts like analysis work. Maybe maybe we can just kind of transition a little bit off. But then again, I just I just know that you have, since you have time, and we’re not doing a time of more new modeling. We’ll kind of just take the sprint as like, okay, let’s say
213 00:29:13.340 ⇒ 00:29:16.850 Uttam Kumaran: you have like 4 h there, 4 h here, and then we kind of just like.
214 00:29:16.890 ⇒ 00:29:19.980 Uttam Kumaran: see? Which client requires the most work.
215 00:29:20.120 ⇒ 00:29:20.940 Ryan Luke Daque: Yeah.
216 00:29:22.630 ⇒ 00:29:24.060 Ryan Luke Daque: Sounds sounds good.
217 00:29:26.260 ⇒ 00:29:28.530 Any other questions?
218 00:29:31.010 ⇒ 00:29:37.969 Ryan Luke Daque: Yeah, I think I’m good. So I guess, for now just continue with the the pool parts questions. I’ll see what I can do.
219 00:29:39.190 ⇒ 00:29:45.739 Ryan Luke Daque: from there and then, maybe later to today or tomorrow, we can discuss more on like, what? What
220 00:29:46.240 ⇒ 00:29:52.810 Ryan Luke Daque: we can work on for what I can work on for acid link. Okay? Great
221 00:29:53.250 ⇒ 00:29:53.970 Ryan Luke Daque: cool.
222 00:29:54.510 ⇒ 00:29:58.680 Uttam Kumaran: Yeah. Hopefully, this will be a lot more modeling work. So
223 00:29:59.470 ⇒ 00:30:00.360 Ryan Luke Daque: nice.
224 00:30:01.180 ⇒ 00:30:02.100 Ryan Luke Daque: Okay.
225 00:30:02.210 ⇒ 00:30:04.530 Uttam Kumaran: sounds good. Slack me, if anything.
226 00:30:04.590 ⇒ 00:30:06.630 Ryan Luke Daque: Sure have a nice rest of your day.
227 00:30:06.720 ⇒ 00:30:08.799 Ryan Luke Daque: Thanks with them. Bye, bye.