Meeting Title: Brainforge Interview w- Demilade Date: 2026-03-06 Meeting participants: Sai Sindhura Poosarla, Zoran Selinger, Demilade Agboola
WEBVTT
1 00:09:41.580 ⇒ 00:09:42.660 Zoran Selinger: Hi, Sai.
2 00:09:48.760 ⇒ 00:10:00.890 Zoran Selinger: Hi, hi. Yeah, so, let me just, tell you, I’m just gonna listen, listen in. So, Demi, you’re having an interview with, with Demi. I’m just gonna sit in,
3 00:10:00.890 ⇒ 00:10:04.880 Sai Sindhura Poosarla: Okay. I’m gonna start interviewing also pretty soon, so I just…
4 00:10:04.960 ⇒ 00:10:07.860 Zoran Selinger: wanna… Wanna see.
5 00:10:07.980 ⇒ 00:10:12.029 Zoran Selinger: a little bit, so… don’t mind me, you just talk to Demi.
6 00:10:12.840 ⇒ 00:10:14.880 Sai Sindhura Poosarla: No worries, nice meeting you, Zoran.
7 00:10:15.170 ⇒ 00:10:16.220 Zoran Selinger: By spending a ton.
8 00:10:19.800 ⇒ 00:10:28.680 Sai Sindhura Poosarla: Hi, Damien. I… for some reason, I can’t hear.
9 00:10:28.680 ⇒ 00:10:29.860 Demilade Agboola: I’m already in here.
10 00:10:30.260 ⇒ 00:10:31.750 Sai Sindhura Poosarla: Yeah, I can hear you now. Hello.
11 00:10:32.340 ⇒ 00:10:36.140 Demilade Agboola: Hi, so I’m Dimladeh, and I was asking, how do you… how do you pronounce your name?
12 00:10:36.400 ⇒ 00:10:38.079 Sai Sindhura Poosarla: You can call me Saek.
13 00:10:38.270 ⇒ 00:10:40.020 Demilade Agboola: Sai, okay, nice to meet you, Sai.
14 00:10:40.480 ⇒ 00:10:41.990 Demilade Agboola: Nice to meet you, Jimmy.
15 00:10:43.600 ⇒ 00:10:48.190 Demilade Agboola: So, I know this is for the analytics engineering role.
16 00:10:49.160 ⇒ 00:10:49.890 Sai Sindhura Poosarla: Yeah.
17 00:10:49.890 ⇒ 00:10:56.099 Demilade Agboola: Yeah, and I know that you have, like, strong interest in that, so we’re gonna… Start off there.
18 00:10:56.200 ⇒ 00:10:59.860 Sai Sindhura Poosarla: Okay. So please, can you tell me about yourself and your experience, and…
19 00:11:00.750 ⇒ 00:11:11.110 Sai Sindhura Poosarla: Yep, for sure, yeah. First of all, thank you for your time, both of you. So I’m Sai, I have about, like, 10 years of experience in the data and analytics field,
20 00:11:11.580 ⇒ 00:11:15.729 Sai Sindhura Poosarla: So, I started off working for Accenture, that was my first job.
21 00:11:15.730 ⇒ 00:11:21.900 Demilade Agboola: I worked there for around 4 years, worked for different clients, Shell, and…
22 00:11:22.000 ⇒ 00:11:35.899 Sai Sindhura Poosarla: I mean, back that time, it was called EMC Square, which was later acquired by Dell. I don’t know how many of, like, people know it, but it was… when I was working, it was called EMC Square, but, I mean, it’s ultimately Dell, now.
23 00:11:36.080 ⇒ 00:11:44.899 Sai Sindhura Poosarla: So there, I started working on, you know, the building SQL queries, pipelines, and, you know, basic report automation and stuff.
24 00:11:45.130 ⇒ 00:11:48.029 Sai Sindhura Poosarla: when I moved to Shell, I did a lot about,
25 00:11:48.240 ⇒ 00:11:58.759 Sai Sindhura Poosarla: you know, the visualization, so I helped the shell company launch a credit card in one of the countries, so it operates in 20 countries.
26 00:11:58.760 ⇒ 00:12:13.530 Sai Sindhura Poosarla: So I helped them launch the credit card policy and the visualizations about the sales and stuff. After getting a little bit of experience in analytics, I moved, to, United States, and I did my master’s.
27 00:12:13.560 ⇒ 00:12:22.740 Sai Sindhura Poosarla: I did my master’s in University of Washington. While I was doing my master’s, I also did get an opportunity to work with,
28 00:12:22.970 ⇒ 00:12:35.880 Sai Sindhura Poosarla: you know, a couple of local healthcare organizations, and those healthcare organizations, like, literally had no idea how to use their data. They had a lot of data, but they don’t know what is data analytics, they don’t know what is data visualization.
29 00:12:35.880 ⇒ 00:12:47.849 Sai Sindhura Poosarla: So, the interesting fact was both of them were healthcare, and I was working, like, at the similar time frame, and I had them set, you know, basic, the dashboards and, stuff for them.
30 00:12:47.850 ⇒ 00:13:02.960 Sai Sindhura Poosarla: For MultiCare, I did also get an opportunity to do a little bit of advanced, data modeling. So, I did create a predictive model that kind of predicted, you know, the probability of an emergency visit to come back.
31 00:13:03.260 ⇒ 00:13:04.490 Sai Sindhura Poosarla: So… Agreed.
32 00:13:04.790 ⇒ 00:13:11.240 Sai Sindhura Poosarla: And after that, I worked for user testing, which is, like, a year and few months.
33 00:13:11.240 ⇒ 00:13:26.989 Sai Sindhura Poosarla: So, it was a startup company, with a small number of team and, less people. That’s the best part of my career. I played, like, it’s just a core analytics team, you know, we were, like, 5 to 6 people, 5 to 6 analysts, like, there was, like, a manager and stuff, but…
34 00:13:26.990 ⇒ 00:13:34.450 Sai Sindhura Poosarla: We got to do all sorts of analytics, like product analytics, marketing analytics, you know, headcount analytics.
35 00:13:34.450 ⇒ 00:13:44.699 Sai Sindhura Poosarla: I mean, everything, there was no end, like, also the data science, so there was nothing like a set boundaries that I should do this, I shouldn’t do this, so I really enjoyed that role.
36 00:13:44.920 ⇒ 00:13:47.909 Demilade Agboola: Alright. Is that I came… I moved to Meta.
37 00:13:48.420 ⇒ 00:14:06.619 Sai Sindhura Poosarla: So from the last 4 years, I’ve been with Meta. So, with Meta, I’m working in the headcount strategy workforce analytics space, so it’s all about, you know, the headcount metrics, like attention, sorry, attrition, retention, you know, people movement.
38 00:14:06.850 ⇒ 00:14:24.500 Sai Sindhura Poosarla: You know, all such fun. So, yeah, I really gained a lot of, you know, deep expertise and depth knowledge about, workforce analytics being with Meta. So, I would say overall, I did get to try a lot of analytics, so, yeah.
39 00:14:25.340 ⇒ 00:14:30.970 Demilade Agboola: Yeah, that’s a lot of experience. You’ve definitely, like, you know, done a lot of things around
40 00:14:31.100 ⇒ 00:14:32.510 Demilade Agboola: Different spaces.
41 00:14:32.650 ⇒ 00:14:34.000 Sai Sindhura Poosarla: Yeah.
42 00:14:34.990 ⇒ 00:14:39.530 Demilade Agboola: What would you say your technical stack is? Like, what tools do you use, and…
43 00:14:40.620 ⇒ 00:14:57.219 Sai Sindhura Poosarla: Currently, I would say, with my… I mean, recently I’ve been with Meta, so my, I use Tableau quite a bit for data visualization, the SQL, but again, like, the internal, there are also a couple of internal visualization tools that I use.
44 00:14:57.270 ⇒ 00:15:12.190 Sai Sindhura Poosarla: And for data engineering or, like, automation of, like, you know, pipeline scheduling, like, it’s, again, like, internal tools. So, I would say this is, like, a tech stack. Obviously, Python for some of the automation or, like, you know, scripting whenever it is needed.
45 00:15:13.840 ⇒ 00:15:17.140 Sai Sindhura Poosarla: But I would, say, like, this used to be…
46 00:15:17.240 ⇒ 00:15:26.960 Sai Sindhura Poosarla: This used to be the tech stack, but then, now we leverage AI for a lot of productivity, so I would say, like,
47 00:15:26.960 ⇒ 00:15:44.030 Sai Sindhura Poosarla: like, a lot has been automated, and there is a lot of productivity increase, using, you know, the AI tools, so I would say, like, we use, Claude, pretty well. And, yeah, and again, like, internal, AI tools developed for different purposes.
48 00:15:44.030 ⇒ 00:15:45.200 Sai Sindhura Poosarla: So, yeah.
49 00:15:46.130 ⇒ 00:15:47.640 Demilade Agboola: Okay, that’s fair, that’s fair.
50 00:15:49.150 ⇒ 00:15:50.709 Demilade Agboola: Okay, I hear that.
51 00:15:50.930 ⇒ 00:15:55.000 Demilade Agboola: I think, from an analytics… perspective.
52 00:15:56.530 ⇒ 00:16:00.120 Demilade Agboola: Do you… on an extension perspective, do you know how to use dbt?
53 00:16:00.210 ⇒ 00:16:22.620 Sai Sindhura Poosarla: That’s a great question. So I’ll be honest with you, so I… when I was back in user testing, I did use dbt to set that, process up, and, like, we also did the migration to Snowflake, so there was, like, a lot of database architecture and setting up stuff with dbt. But to be very honest, when I moved to Meta, I did not get to use that exact tool.
54 00:16:22.620 ⇒ 00:16:29.260 Sai Sindhura Poosarla: But I still set up the data architecture, or pipelines, or using Python scripting, or… In a different format.
55 00:16:29.310 ⇒ 00:16:35.759 Sai Sindhura Poosarla: Yeah, but not the exact tool, because, yeah, I mean, yeah, I don’t have control on what tools I should use, yeah.
56 00:16:36.450 ⇒ 00:16:40.130 Demilade Agboola: Yeah, that’s fair, that’s fair. I just wanted to know if you’ve had experience with DBT? Yes, yes.
57 00:16:40.130 ⇒ 00:16:42.840 Sai Sindhura Poosarla: Our modeling experience is always very important.
58 00:16:43.090 ⇒ 00:16:44.280 Sai Sindhura Poosarla: Yep. Okay.
59 00:16:46.050 ⇒ 00:16:50.979 Demilade Agboola: Alright, so, like, let’s start off thinking, like, system architecture questions, and just…
60 00:16:50.980 ⇒ 00:16:51.610 Sai Sindhura Poosarla: Yep.
61 00:16:51.610 ⇒ 00:16:54.370 Demilade Agboola: Quick, high-level questions.
62 00:16:56.310 ⇒ 00:17:03.840 Demilade Agboola: So, if you have… raw data coming in. What are the tests you would want to do and ensure
63 00:17:04.160 ⇒ 00:17:09.290 Demilade Agboola: That the data has, before we start, like, modeling that data.
64 00:17:11.069 ⇒ 00:17:20.179 Sai Sindhura Poosarla: Yeah, that’s a good question. So, I think there are different ways to handle this. Like, first of all, I want to understand what is the end goal, and what is the end
65 00:17:20.339 ⇒ 00:17:26.489 Sai Sindhura Poosarla: product that we want to make from this. If… if we have that.
66 00:17:26.639 ⇒ 00:17:46.259 Sai Sindhura Poosarla: then well and good, I would start from there. But if it is, like, an open-ended question, okay, here is the data, how do I use it? Then I think the initial thing is I would like to understand the granularity, and there is… if there is a unique identifier, right? How do I define a unique identifier or a primary key?
67 00:17:46.299 ⇒ 00:17:50.809 Sai Sindhura Poosarla: for… For a specific raw data coming in.
68 00:17:51.079 ⇒ 00:18:00.079 Sai Sindhura Poosarla: Okay. If… yeah, so that is one thing, making sure there are no, like, lot of nulls, making sure, like, we have…
69 00:18:00.209 ⇒ 00:18:20.119 Sai Sindhura Poosarla: you know, data integrity. Like, if there are, like, 2 or 3 files that are needed to be converted into, tables, or, like, established that data modeling, then I would say, like, understanding the data connections between the files. Do I have the same… if I have the customer ID here, do I have the customer ID here, or not? Otherwise, it’s very hard.
70 00:18:20.119 ⇒ 00:18:29.029 Sai Sindhura Poosarla: To understand the mapping, So those are the couple of things, and what else?
71 00:18:29.619 ⇒ 00:18:45.099 Sai Sindhura Poosarla: basically understanding the data type of each column, like, if the data type… if that is right or wrong. If not, it has to be transformed to the way it needs to be fit into our current table structure.
72 00:18:45.980 ⇒ 00:18:46.899 Demilade Agboola: Alright, that’s fair.
73 00:18:47.100 ⇒ 00:18:51.020 Demilade Agboola: Subsequent questions would be…
74 00:18:52.980 ⇒ 00:18:58.419 Demilade Agboola: how do you structure your modeling? So, what I mean by that is, number one.
75 00:18:58.750 ⇒ 00:19:08.099 Demilade Agboola: how do you do your joins? Like, in terms of, like, when you’re trying to build, what’s the end goal? How do you try to go from your raw tables to your mat?
76 00:19:08.600 ⇒ 00:19:17.440 Demilade Agboola: And two is what type of technique in terms of data modeling? So, star schemas, in terms of normalized schemas, like, how do you go about that?
77 00:19:18.110 ⇒ 00:19:37.019 Sai Sindhura Poosarla: Okay, perfect. Yeah, I think first, I believe, like, we need to understand, like, how to build the dimension tables, like, the raw data tables, and then it goes to the aggregate tables. So for this, yeah, for this, I would, kind of use, like, yeah, different schemas available. Like, I would say, like, the star schema.
78 00:19:37.090 ⇒ 00:19:40.519 Sai Sindhura Poosarla: Where it is, like, you know, you have,
79 00:19:41.480 ⇒ 00:19:47.229 Sai Sindhura Poosarla: You have the basic dimension tables, and then you kind of connect it to the aggregate table, so…
80 00:19:47.490 ⇒ 00:19:50.940 Sai Sindhura Poosarla: We could use different, schemas available, yeah.
81 00:19:53.660 ⇒ 00:19:56.440 Demilade Agboola: And then, in terms of just being able to…
82 00:19:56.740 ⇒ 00:20:07.079 Demilade Agboola: handle, like, data observability? How do you ensure that the data is… Of high quality consistently.
83 00:20:08.500 ⇒ 00:20:15.470 Sai Sindhura Poosarla: So the data is of high quality. So you mean the incoming data, or, like, the.
84 00:20:15.470 ⇒ 00:20:19.429 Demilade Agboola: So for the end users. So, you know, you model all this data.
85 00:20:20.070 ⇒ 00:20:22.889 Demilade Agboola: How do you ensure that what goes out to the…
86 00:20:22.890 ⇒ 00:20:24.470 Sai Sindhura Poosarla: business stakeholders.
87 00:20:24.470 ⇒ 00:20:27.800 Demilade Agboola: It’s of high quality, so, like, we ensure that,
88 00:20:28.010 ⇒ 00:20:29.190 Sai Sindhura Poosarla: Got you, yeah.
89 00:20:29.190 ⇒ 00:20:32.720 Demilade Agboola: You know, freshness, for instance. We ensure that.
90 00:20:32.970 ⇒ 00:20:52.099 Sai Sindhura Poosarla: Yeah, I would say, like, we can set up some, validation checks, like, making sure the data is refreshing, how, like, on the point that you touched on the freshness, like, how, like, what is the frequency of data update? Is it, like, hourly? Is it daily, depending upon the importance of the data updates?
91 00:20:52.380 ⇒ 00:21:01.689 Sai Sindhura Poosarla: And the other thing is we, like, the best way to understand if the data is accurate or not is to set up the validation checks, right?
92 00:21:01.690 ⇒ 00:21:18.049 Sai Sindhura Poosarla: like, overall, like, okay, I need to get, you know, these many records around for this day, then definitely, like, seeing the pattern, like, every day I get, like, you know, 1 million records, like, today, did I get 5 million records? Is it really right, or do I… am I getting any duplicates?
93 00:21:18.080 ⇒ 00:21:30.209 Sai Sindhura Poosarla: So, setting up such validation checks, which are more, like, data-related, and some of the validation checks could be, you know, business or context-related, you know.
94 00:21:30.370 ⇒ 00:21:43.919 Sai Sindhura Poosarla: Yeah, some checks could be, you know, sometimes the, you know, you have sales, but you get, like, you know, sales ID or, like, some sales customer name in that. So, yeah, I mean, we could go, like, as comprehensive as possible with validation checks.
95 00:21:44.050 ⇒ 00:21:50.949 Sai Sindhura Poosarla: To put the validation checks on the end column, and we could do that monitoring for a couple of
96 00:21:50.990 ⇒ 00:22:08.749 Sai Sindhura Poosarla: weeks or so, and, like, the way I used to do is, like, set those, validation checks and send that email to the working team, so that whoever has time, they could go and look at the issues and fix it. Once we… once I feel like it’s more, stabilized and things are stable.
97 00:22:08.750 ⇒ 00:22:14.360 Sai Sindhura Poosarla: Then I would, just say, like, the validation checks probably, like, from daily to…
98 00:22:14.360 ⇒ 00:22:20.849 Sai Sindhura Poosarla: Weekly, reduce the frequency, so it depends upon when the project actually stabilizes.
99 00:22:21.480 ⇒ 00:22:22.210 Demilade Agboola: Oh, okay.
100 00:22:23.700 ⇒ 00:22:28.109 Demilade Agboola: My question here would be…
101 00:22:28.230 ⇒ 00:22:33.479 Demilade Agboola: How do you build up models, or how do you ensure that you’re building the right models?
102 00:22:36.730 ⇒ 00:22:42.540 Demilade Agboola: And how do you, like, how do you ensure you’re building the right models? How do you gain clarity from the stakeholders? How do you ensure that…
103 00:22:42.690 ⇒ 00:22:48.639 Demilade Agboola: You don’t want to spend 4 hours, 5 hours building models that have no business utility.
104 00:22:49.440 ⇒ 00:22:59.930 Sai Sindhura Poosarla: Yep, that’s a great question, so I think this goes back to, like, when I get the data, I think it’s about, like, understanding the requirements from the stakeholders, documenting it, and making sure that, you know.
105 00:22:59.970 ⇒ 00:23:14.799 Sai Sindhura Poosarla: this is what they want. So before I put in, like, a lot of effort, do everything, I think I could use prototyping to show them, like, this is how I… this is my plan, I’m going to do this. At the end, this would be the result.
106 00:23:14.840 ⇒ 00:23:22.069 Sai Sindhura Poosarla: You know, it could be a mix of technical and non-technical folks, so depending upon who my business is.
107 00:23:22.120 ⇒ 00:23:30.589 Sai Sindhura Poosarla: If it is, like, all… it’s, like, pretty high-level people, like senior executives, I wouldn’t really get into the design, it would be, like, a one slide.
108 00:23:30.740 ⇒ 00:23:50.049 Sai Sindhura Poosarla: like, overall, it would be, like, a high-level, you know, presentation, but if I’m talking to other data engineers or other, you know, individual contributors, then yes, I would definitely love to get into the details, get the sign-off alignment on, this is how I’m going to do, and if
109 00:23:50.180 ⇒ 00:23:55.100 Sai Sindhura Poosarla: this completely satisfies this purpose or not, and then I would start
110 00:23:55.270 ⇒ 00:24:04.739 Sai Sindhura Poosarla: building stuff, but the prototyping or, like, getting the minimal viable product, understanding the requirements, I would say these are a couple of things.
111 00:24:05.530 ⇒ 00:24:06.180 Demilade Agboola: Okay.
112 00:24:10.410 ⇒ 00:24:15.880 Demilade Agboola: Okay, fair enough. I think the other thing I would ask is… Because I know…
113 00:24:16.060 ⇒ 00:24:24.379 Demilade Agboola: like, I work, obviously working forge, and sometimes the AEs still have to do some analyst work, because, you know, we’re a small team.
114 00:24:24.610 ⇒ 00:24:31.650 Demilade Agboola: So I think my question would be around that, so what tools can you use in that sense? I know you mentioned Tableau.
115 00:24:32.230 ⇒ 00:24:33.940 Demilade Agboola: Any other tools you can use.
116 00:24:34.920 ⇒ 00:24:37.040 Sai Sindhura Poosarla: For data analysis.
117 00:24:37.230 ⇒ 00:24:37.780 Demilade Agboola: Yes.
118 00:24:38.120 ⇒ 00:24:40.449 Demilade Agboola: Or, like, building out dashboards.
119 00:24:40.650 ⇒ 00:24:41.710 Sai Sindhura Poosarla: Yeah
120 00:24:41.850 ⇒ 00:24:50.839 Sai Sindhura Poosarla: there are, like, a wide variety of data visualization tools, so, like, there is Looker, there is Tableau, there is, like,
121 00:24:50.840 ⇒ 00:25:02.989 Sai Sindhura Poosarla: the Power BI from Microsoft, and… I mean, these are all licensed. We also have, like, a couple of other tools that are not licensed as well, like, which I think I need to do some research, but…
122 00:25:02.990 ⇒ 00:25:03.580 Demilade Agboola: Oh, okay.
123 00:25:03.580 ⇒ 00:25:12.310 Sai Sindhura Poosarla: there was, like, Periscope long back when I used it, but I don’t know if it has been… I think it was renamed to,
124 00:25:12.490 ⇒ 00:25:20.510 Sai Sindhura Poosarla: Simon or something, like, I don’t exactly remember, so there are a couple of other versions as well, but if it is not tool.
125 00:25:20.510 ⇒ 00:25:33.900 Sai Sindhura Poosarla: I just feel like plain SQL could also be used to do the data analysis and extract the data and use a basic, you know, Gsheet to create some visualizations and
126 00:25:34.410 ⇒ 00:25:50.270 Sai Sindhura Poosarla: I feel like the simpler the visualization is, the more the audience would connect, rather than making fancy visualizations. If I make a plain line chart or, like, a bar graph that has proper recommendations that improves the business.
127 00:25:50.370 ⇒ 00:26:00.550 Sai Sindhura Poosarla: That is… that would be well received than, making, like, a lot of complex stuff and, you know, the business really don’t understand what they have to take.
128 00:26:00.660 ⇒ 00:26:18.199 Sai Sindhura Poosarla: So I would say, like, you could use, like, as simple as a Gsheet, or you could go fancy with visualization tools, or I would say, like, in R and in Python as well, like, we have inbuilt libraries to get those visualizations, so we could also do that, and…
129 00:26:18.850 ⇒ 00:26:27.130 Sai Sindhura Poosarla: Yeah, and if we have… if you’re using AI, then, yeah, I mean, there’s no end, yeah. We could go, like, we could do anything, yeah.
130 00:26:28.090 ⇒ 00:26:37.499 Demilade Agboola: Okay, fair enough, fair enough. No, so the reason why I ask that is because, I mean, I want to know the tools you’re comfortable using. I know you mentioned Tableau, I know you mentioned Looker.
131 00:26:37.550 ⇒ 00:26:39.359 Sai Sindhura Poosarla: Yes. And Power BI.
132 00:26:39.360 ⇒ 00:26:41.909 Demilade Agboola: Yes. How comfortable are you with, like, LookML?
133 00:26:42.680 ⇒ 00:26:44.370 Sai Sindhura Poosarla: Sorry, LookML.
134 00:26:44.370 ⇒ 00:26:44.990 Demilade Agboola: Yes.
135 00:26:46.160 ⇒ 00:26:58.070 Sai Sindhura Poosarla: So I think I’ve not worked with it, but if it’s a data visualization tool with the experience that I have, I think I should be able to pick it up. Should be lesser onboarding, yeah.
136 00:26:58.520 ⇒ 00:27:04.639 Demilade Agboola: Yeah, okay, so I… so… I mean, like, LookML is kind of, like, within Looker, in the sense of…
137 00:27:06.070 ⇒ 00:27:09.910 Demilade Agboola: It’s the SQL semantic layer that you build out.
138 00:27:10.020 ⇒ 00:27:16.739 Demilade Agboola: within, like, look, I was just asking these questions because I personally…
139 00:27:17.100 ⇒ 00:27:19.730 Demilade Agboola: I’ve had to, like, dabble with some tools.
140 00:27:20.680 ⇒ 00:27:26.609 Demilade Agboola: Gotcha. Because I’ll start setting projects, and I just want to know how comfortable you are with, like, you know, dabbling.
141 00:27:26.610 ⇒ 00:27:40.059 Sai Sindhura Poosarla: Yeah, the recent tech stack I’ve used is Tableau. The Looker and Periscope, and I used in user testing, and Power BI in my earlier internships. So, yeah, the recent one I used is Tableau.
142 00:27:42.290 ⇒ 00:27:45.479 Demilade Agboola: Okay, sure, that sounds good.
143 00:27:45.940 ⇒ 00:27:47.600 Demilade Agboola: Okay, in terms of…
144 00:27:51.310 ⇒ 00:27:55.670 Demilade Agboola: like, dbt, I know you mentioned SQL,
145 00:27:56.380 ⇒ 00:28:05.439 Demilade Agboola: how comfortable are you with, especially because you’ve not used DVC in a while, and the number of changes, but how comfortable are you with, like, learning new things and new techniques?
146 00:28:05.590 ⇒ 00:28:09.809 Demilade Agboola: And just being able to pick that up, just so that, you know…
147 00:28:10.820 ⇒ 00:28:14.290 Demilade Agboola: for, like, 4 years now, so I just wanted to know how comfortable you are with.
148 00:28:14.760 ⇒ 00:28:15.730 Sai Sindhura Poosarla: Yeah.
149 00:28:15.960 ⇒ 00:28:35.479 Sai Sindhura Poosarla: Yeah, I definitely feel things would have changed quite a bit from the time I have used DPT till now. There would have been, like, a lot of enhancements and updates to the tool as well. But if I have to learn something new, I think I… I would definitely understand, easily, quickly, and I always love and, you know, explore
150 00:28:35.480 ⇒ 00:28:49.869 Sai Sindhura Poosarla: led to explore things, and also, like, with AI, like, every day we get a new tool, so I think it… it is, like, a… it has to be, like, a mandatory thing for us to, you know, keep learning new stuff as it gets launched.
151 00:28:49.900 ⇒ 00:28:52.869 Sai Sindhura Poosarla: So, also, I’m more of a person
152 00:28:53.320 ⇒ 00:29:07.649 Sai Sindhura Poosarla: like, who is more hands-on, rather than tutorials. So, if I get any tool that I need to learn, I would just, like, directly jump in, try to do stuff by myself, take a small scenario or, like, a small use case.
153 00:29:07.720 ⇒ 00:29:21.049 Sai Sindhura Poosarla: like, yeah, just get, like, sales by 4 years, like, if I get a new tool, like, how to do that, or, like, I would just start exploring by myself, along with understanding tutorials a little bit more,
154 00:29:21.050 ⇒ 00:29:31.219 Sai Sindhura Poosarla: about the tool. And also, like, there is always, like, a lot to learn about the tool, but I would definitely narrow it down to what is my immediate deliverable.
155 00:29:31.220 ⇒ 00:29:38.590 Sai Sindhura Poosarla: And I would like to focus on those features instead of, you know, just going crazy on learning everything possible with the tool.
156 00:29:39.530 ⇒ 00:29:40.839 Demilade Agboola: Okay, fair enough, fair enough.
157 00:29:44.020 ⇒ 00:29:50.599 Demilade Agboola: I think, in terms of… I know we talked about, like, star schemas and normalized schemas.
158 00:29:51.230 ⇒ 00:29:56.089 Demilade Agboola: Curious as to when do you prefer a star schema and when do you prefer a normalized schema?
159 00:29:57.300 ⇒ 00:29:59.969 Sai Sindhura Poosarla: When do I prefer a star?
160 00:29:59.970 ⇒ 00:30:05.450 Demilade Agboola: Okay, so when would you consider the use case for a star schema versus the use case for a normalized schema?
161 00:30:06.880 ⇒ 00:30:14.350 Sai Sindhura Poosarla: Okay. Star schema versus normalized schema… Okay, when I…
162 00:30:20.310 ⇒ 00:30:27.760 Sai Sindhura Poosarla: Yeah, I think normalization is more, when we want to have, like.
163 00:30:28.500 ⇒ 00:30:45.299 Sai Sindhura Poosarla: Yeah, I think recently we kind of used a lot of normalized schema, right? Like, star schema was, like, kind of used in, like, a little bit olderen days. Like, normalized schema is, like, where we have dimension tables for every, every… what is that? Every component.
164 00:30:45.610 ⇒ 00:30:52.719 Sai Sindhura Poosarla: And a star schema is more like, we have aggregate stuff, and then, like, you kind of do that. Yeah.
165 00:30:53.050 ⇒ 00:31:02.560 Demilade Agboola: Yeah, I mean, to be fair, it’s kind of, like, it’s flipped, but it’s the same thing, like, so star schemas are when you have, like, your fact tables and your dimensional tables, and you kind of put them together.
166 00:31:02.770 ⇒ 00:31:08.609 Demilade Agboola: And then your normalized schemas are, like, when you have, like, aggregate, and you try to roll up your data, and…
167 00:31:09.280 ⇒ 00:31:11.009 Demilade Agboola: Into different things.
168 00:31:11.010 ⇒ 00:31:11.600 Sai Sindhura Poosarla: Hmm.
169 00:31:14.100 ⇒ 00:31:15.890 Demilade Agboola: Yeah, so I, I, I think…
170 00:31:16.080 ⇒ 00:31:19.430 Demilade Agboola: So, my question would be, like, when would you want to use a star schema?
171 00:31:19.570 ⇒ 00:31:23.080 Demilade Agboola: And when would you… Prefer to use, like, a normalized schema.
172 00:31:29.420 ⇒ 00:31:48.490 Sai Sindhura Poosarla: I would just say, depending upon the use case, so if they just, have, like, only the aggregate data, probably the star schema, but if I, like, use, like, the normalized one, like, if I have to go really deep into understanding all the data and getting the insights, then I would go for the normalized one.
173 00:31:49.810 ⇒ 00:31:50.429 Demilade Agboola: Okay.
174 00:31:53.330 ⇒ 00:31:55.260 Demilade Agboola: Okay, that’s fair. I…
175 00:31:55.260 ⇒ 00:31:58.589 Sai Sindhura Poosarla: I think we’re almost on time, we have, like, 8 minutes left.
176 00:31:58.590 ⇒ 00:32:04.549 Demilade Agboola: I’m personally out of questions. Do you have any questions about, like, Brainforge or, like, the role?
177 00:32:05.330 ⇒ 00:32:13.060 Sai Sindhura Poosarla: Yeah, so my question is, like, what are the recent, projects that you have dealt with, and what are the challenges you faced, or, like, yeah.
178 00:32:15.770 ⇒ 00:32:18.489 Demilade Agboola: A lot of projects tend to be…
179 00:32:19.150 ⇒ 00:32:24.800 Demilade Agboola: so my recent projects, I have worked on helping to migrate, Cross…
180 00:32:25.790 ⇒ 00:32:34.780 Demilade Agboola: Helping the client migrate data from Tableau between you, visualization called Omni.
181 00:32:35.010 ⇒ 00:32:35.370 Sai Sindhura Poosarla: I’m gonna…
182 00:32:35.370 ⇒ 00:32:41.380 Demilade Agboola: Because certain ways we modeled were for Tableau, the modeling just didn’t transfer over to Omni, it would break.
183 00:32:41.680 ⇒ 00:32:45.759 Demilade Agboola: So having to, like, make those fixes and ensure that how we
184 00:32:45.950 ⇒ 00:32:54.059 Demilade Agboola: Remodeling what had already remodeled so that it would work in Omni was number one. Another thing I have been working on recently…
185 00:32:54.280 ⇒ 00:33:03.800 Demilade Agboola: has been… A new client that we have, who’s just, like, moving… Everything, basically, into, like.
186 00:33:04.040 ⇒ 00:33:05.340 Demilade Agboola: A warehouse now.
187 00:33:05.610 ⇒ 00:33:12.720 Demilade Agboola: So just being responsible for, like, ingesting the data, modeling the data, and building out the.
188 00:33:12.720 ⇒ 00:33:13.420 Sai Sindhura Poosarla: Surely.
189 00:33:13.420 ⇒ 00:33:16.600 Demilade Agboola: in the dashboards as well.
190 00:33:18.240 ⇒ 00:33:25.529 Demilade Agboola: And then, yes, I have another client as well, I’m modeling some of their data that we’re ingesting from an API called Spins.
191 00:33:25.660 ⇒ 00:33:34.890 Demilade Agboola: So we’re interested in API data in there. So I wasn’t handling the ingestion, another teammate of mine was, but I was responsible for the modeling.
192 00:33:35.220 ⇒ 00:33:36.440 Sai Sindhura Poosarla: of that data.
193 00:33:36.470 ⇒ 00:33:38.979 Demilade Agboola: There’s also, like, you know, base models.
194 00:33:39.490 ⇒ 00:33:40.240 Sai Sindhura Poosarla: Interests.
195 00:33:40.240 ⇒ 00:33:45.919 Demilade Agboola: As well as, like, making it… ingesting, like, creating, like, Mods that you can see.
196 00:33:46.370 ⇒ 00:33:47.409 Demilade Agboola: by week?
197 00:33:47.590 ⇒ 00:33:56.089 Demilade Agboola: the values for the different… because it wasn’t just spins, I was integrating spins with the previous data as well, so just seeing the values for that as well.
198 00:33:57.050 ⇒ 00:33:59.979 Sai Sindhura Poosarla: Okay, yeah, that sounds great. Sure, yeah.
199 00:34:00.710 ⇒ 00:34:05.989 Demilade Agboola: So, yeah, a lot of work here, at least from an AE perspective, revolves around, like.
200 00:34:07.170 ⇒ 00:34:13.260 Demilade Agboola: modeling… But, like, that’s why I also asked you questions around how comfortable you are with BI tools, because…
201 00:34:13.690 ⇒ 00:34:22.139 Demilade Agboola: Especially when the project hasn’t gotten to the stage where we will have, like, a full-on analyst on the team to just be charting our dashboard.
202 00:34:23.330 ⇒ 00:34:25.070 Demilade Agboola: Proof of concepts and the things that you.
203 00:34:25.070 ⇒ 00:34:25.460 Sai Sindhura Poosarla: Interesting.
204 00:34:25.469 ⇒ 00:34:26.629 Demilade Agboola: QA the data.
205 00:34:27.699 ⇒ 00:34:37.980 Demilade Agboola: dashboards you’ll build off of it, so you can show all revenue over the last 12 months, for instance. And then the stakeholder can look at it and go, yes, this looks about right, or no, this just seems off.
206 00:34:38.090 ⇒ 00:34:42.659 Demilade Agboola: And so you can go back to the models, remodel it, and, you know, show them again as proof of concept.
207 00:34:43.449 ⇒ 00:34:51.019 Sai Sindhura Poosarla: Yeah, I think that sounds great, yeah. What are the general challenges that you face, like, when you deal with,
208 00:34:51.139 ⇒ 00:35:04.449 Sai Sindhura Poosarla: different set of clients, or, like, how does the day-to-day work look? Is it, like, you focus on one client at a time, or, like, you have, like, a couple of projects running in parallel? Like, how does it work at Brainforge?
209 00:35:05.530 ⇒ 00:35:10.550 Demilade Agboola: Depends on the person, depends on the scope, depends on, like, work.
210 00:35:10.680 ⇒ 00:35:12.829 Demilade Agboola: Hours that you have.
211 00:35:13.290 ⇒ 00:35:13.910 Sai Sindhura Poosarla: Right.
212 00:35:13.910 ⇒ 00:35:21.489 Demilade Agboola: Because some people are part-time, so obviously they’re not going to have as much scope. Some are full-time, so they have much more scope.
213 00:35:22.230 ⇒ 00:35:26.490 Demilade Agboola: But ultimately, The idea is…
214 00:35:27.630 ⇒ 00:35:36.169 Demilade Agboola: you will be on usually a couple, maybe two projects, depending on the weight of the project. So some projects you might be in, like.
215 00:35:36.550 ⇒ 00:35:38.099 Demilade Agboola: a supporting role.
216 00:35:38.450 ⇒ 00:35:39.140 Sai Sindhura Poosarla: Z.
217 00:35:39.140 ⇒ 00:35:52.840 Demilade Agboola: some other products you might be leading. So it’s just, like, finding that right balance, and generally speaking, Europe, like, we maintain, like, an open door policy, so you can always reach out to Utam, who is the CEO.
218 00:35:53.550 ⇒ 00:36:00.440 Demilade Agboola: or the ops team, and just say, hey, like, I feel like my workload is too… is too much, like, of recent.
219 00:36:00.930 ⇒ 00:36:07.640 Demilade Agboola: And if I want to be able to deliver this, this, this to the client, you know, I might need, like, support in this area, or…
220 00:36:07.810 ⇒ 00:36:13.540 Demilade Agboola: I might need someone to fill in for me for, like, this week, or in the next couple of weeks, or whatever you might need.
221 00:36:13.710 ⇒ 00:36:21.200 Demilade Agboola: So in that regard, yes, the… on a day-to-day, you probably will be on, like, more than one.
222 00:36:21.420 ⇒ 00:36:22.490 Demilade Agboola: Maybe two.
223 00:36:22.710 ⇒ 00:36:29.960 Demilade Agboola: potentially 3. Right. But it could not be 3, like, leading. Like, there’s no way you’re going to be leading all 3, because that would be too much on you.
224 00:36:30.270 ⇒ 00:36:30.700 Sai Sindhura Poosarla: Right.
225 00:36:30.700 ⇒ 00:36:39.790 Demilade Agboola: So it would just be, like, figuring out that balance between, like, support, lead, and, like, communicating how you’re finding your experience on a day-to-day to the…
226 00:36:41.030 ⇒ 00:36:41.620 Sai Sindhura Poosarla: Yep.
227 00:36:42.600 ⇒ 00:36:43.390 Demilade Agboola: Oops, too.
228 00:36:44.120 ⇒ 00:36:45.170 Sai Sindhura Poosarla: Okay, yeah.
229 00:36:45.340 ⇒ 00:36:48.369 Sai Sindhura Poosarla: I think that makes sense, yeah. Those are all the questions I had, yeah.
230 00:36:48.370 ⇒ 00:36:51.889 Demilade Agboola: Okay, that’s fine, alright, in that case, I think…
231 00:36:52.660 ⇒ 00:36:57.030 Demilade Agboola: Because we’re almost at time. I think we can call it a day.
232 00:36:58.000 ⇒ 00:37:06.389 Demilade Agboola: So I’ll be… we have a head of recruitment, I’ll reach out to her, give her my thoughts, and she’ll be responsible for what happens next.
233 00:37:06.690 ⇒ 00:37:10.789 Sai Sindhura Poosarla: Sure, yeah, thank you. Thank you for your time. Have a good weekend.
234 00:37:11.170 ⇒ 00:37:12.000 Demilade Agboola: True. Bye.
235 00:37:12.000 ⇒ 00:37:13.379 Zoran Selinger: Nice to meet you, Mike.
236 00:37:13.380 ⇒ 00:37:14.689 Sai Sindhura Poosarla: Bye, Zaran, bye.