Meeting Title: Brainforge Interview w- Demilade Date: 2026-03-12 Meeting participants: Ajibade Adeleke, Demilade Agboola
WEBVTT
1 00:01:12.180 ⇒ 00:01:13.269 Ajibade Adeleke: Coffee sights.
2 00:01:18.900 ⇒ 00:01:22.120 Ajibade Adeleke: Maybe… JoJo.
3 00:03:26.650 ⇒ 00:03:27.530 Demilade Agboola: Alright, GBD.
4 00:03:27.530 ⇒ 00:03:27.850 Ajibade Adeleke: Excellent.
5 00:03:28.320 ⇒ 00:03:30.150 Demilade Agboola: Good afternoon. Give me one minute.
6 00:03:31.100 ⇒ 00:03:31.830 Ajibade Adeleke: Cool.
7 00:03:32.780 ⇒ 00:03:33.620 Ajibade Adeleke: You know what I mean?
8 00:04:05.300 ⇒ 00:04:08.520 Ajibade Adeleke: And then in case you’re… in case if you’re speaking, you’re muted.
9 00:04:13.330 ⇒ 00:04:20.000 Demilade Agboola: I wasn’t actually… give me one second, I just had one minute to get some things ready. Hi.
10 00:04:20.310 ⇒ 00:04:23.360 Demilade Agboola: My name is Damonade, nice to meet you, everybody.
11 00:04:24.980 ⇒ 00:04:28.520 Demilade Agboola: So, I know you’ve had the first call with OH, I believe?
12 00:04:28.870 ⇒ 00:04:31.189 Demilade Agboola: And so, the idea of this…
13 00:04:31.190 ⇒ 00:04:31.950 Ajibade Adeleke: That’s great.
14 00:04:32.890 ⇒ 00:04:33.600 Demilade Agboola: Okay.
15 00:04:34.270 ⇒ 00:04:38.409 Demilade Agboola: So the idea of this second call is just for us to go over…
16 00:04:38.630 ⇒ 00:04:44.620 Demilade Agboola: Certain, like, technical concepts, but, like, first things first, can you please tell me about yourself and your work experience?
17 00:04:47.060 ⇒ 00:04:47.640 Ajibade Adeleke: Purple.
18 00:04:55.130 ⇒ 00:04:55.860 Demilade Agboola: Can you hear me?
19 00:04:56.700 ⇒ 00:04:57.980 Ajibade Adeleke: Yeah, yeah, I got right now.
20 00:04:58.210 ⇒ 00:05:01.920 Demilade Agboola: Yes, I said, please, can you tell me about yourself and your work experience, and what you’ve done?
21 00:05:04.840 ⇒ 00:05:06.859 Ajibade Adeleke: I think you’re fully ongoing.
22 00:05:09.000 ⇒ 00:05:13.720 Demilade Agboola: should be fine, I… Can you, like, can you hear me properly?
23 00:05:13.720 ⇒ 00:05:15.630 Ajibade Adeleke: Yes, yes, I can hear you properly.
24 00:05:15.990 ⇒ 00:05:19.180 Demilade Agboola: So I’m saying… please, can you tell me about yourself and your work experience?
25 00:05:19.930 ⇒ 00:05:21.989 Ajibade Adeleke: Okay, thank you.
26 00:05:22.230 ⇒ 00:05:24.399 Ajibade Adeleke: Let me confirm if you can hear your claim.
27 00:05:25.960 ⇒ 00:05:27.529 Ajibade Adeleke: Oh, it’s,
28 00:05:27.810 ⇒ 00:05:41.579 Ajibade Adeleke: My name is Ajib Alina. I’m a data analytics engineer with over 5 years of experience building end-to-end data systems that basically transform all multi-source data into
29 00:05:42.120 ⇒ 00:06:01.729 Ajibade Adeleke: global reporting layer. So, in my 22 tech, and public sector, you know, from field, from designing ingestion pipeline, to building DVT model layer, and then, delivering the BI dashboard, and basically 40% by 70%.
30 00:06:01.830 ⇒ 00:06:10.010 Ajibade Adeleke: How I like to work is I would… I partner across hospitality across finance, HR,
31 00:06:10.010 ⇒ 00:06:22.960 Ajibade Adeleke: for clients, for questions, or product leaders, to basically translate their questions into a scalable data solution. Now, how I like to work is, I ship the IS deliverables first.
32 00:06:22.960 ⇒ 00:06:36.049 Ajibade Adeleke: I update, proactively update stakeholders, and I’m obsessed with data quality. And some of my part-off achievements is centralizing of a $1.7 billion international claim.
33 00:06:36.220 ⇒ 00:06:39.530 Ajibade Adeleke: Which reduces the concentration by 35%.
34 00:06:39.610 ⇒ 00:06:49.350 Ajibade Adeleke: I’ve also basically built an analytics layer for a what? For a payment platform called Advancedly, which basically,
35 00:06:49.370 ⇒ 00:07:01.670 Ajibade Adeleke: increase their business oversight by 30%, and I’ve also worked on SLA’s dashboard that improves their compliance by 22%.
36 00:07:02.250 ⇒ 00:07:07.630 Ajibade Adeleke: And most recently, I was designated as the theater champion in the OKR Business Coffee World.
37 00:07:10.400 ⇒ 00:07:12.939 Demilade Agboola: That sounds pretty good, that’s actually very impressive.
38 00:07:14.090 ⇒ 00:07:22.230 Demilade Agboola: So, that… we will start off with just, like, walking through some system design questions and trying to understand how you work.
39 00:07:22.800 ⇒ 00:07:33.729 Demilade Agboola: If you had 3 data sources, so you had, Google Ads, you had Stripe.
40 00:07:33.930 ⇒ 00:07:39.790 Demilade Agboola: and you had Shopify, And a stakeholder asks you for a revenue mat.
41 00:07:40.110 ⇒ 00:07:42.429 Demilade Agboola: How would you go about building
42 00:07:42.610 ⇒ 00:07:51.829 Demilade Agboola: your revenue, Matt. So, just walk me through the processes of what you first do, how you handle different, like, the different processes, and what you need to do to go from…
43 00:07:52.310 ⇒ 00:07:56.599 Demilade Agboola: Those data sources to having a data mat.
44 00:07:56.730 ⇒ 00:07:57.959 Demilade Agboola: Ready for stakeholders.
45 00:07:57.960 ⇒ 00:08:06.079 Ajibade Adeleke: Okay, thank you very much. So basically, since I have those three resources, Stripe, Google Analytics, and what did you say, David?
46 00:08:06.410 ⇒ 00:08:06.960 Ajibade Adeleke: Totally.
47 00:08:06.960 ⇒ 00:08:10.979 Demilade Agboola: So Stripe, Google Analytics, and, Shopify.
48 00:08:11.800 ⇒ 00:08:15.820 Ajibade Adeleke: Okay, so I would understand that these are, like, OLTP systems.
49 00:08:15.980 ⇒ 00:08:25.670 Ajibade Adeleke: Right? So, the first thing I would basically want to do, even before diving into the goal, trying to, understand.
50 00:08:26.740 ⇒ 00:08:30.930 Ajibade Adeleke: You know, so… Our basic kind of basics, correct?
51 00:08:31.070 ⇒ 00:08:33.020 Ajibade Adeleke: Sorry.
52 00:08:33.020 ⇒ 00:08:36.570 Demilade Agboola: Your network seems to be breaking, so I’m struggling to pick you out.
53 00:08:39.190 ⇒ 00:08:40.829 Ajibade Adeleke: Yeah, oh no, she’s ringing.
54 00:08:42.049 ⇒ 00:08:42.900 Ajibade Adeleke: Switch.
55 00:08:45.050 ⇒ 00:08:45.839 Ajibade Adeleke: Zip scheme.
56 00:08:51.700 ⇒ 00:08:52.830 Ajibade Adeleke: How do I know.
57 00:08:53.670 ⇒ 00:08:55.100 Demilade Agboola: It’s better. Let’s keep going.
58 00:08:55.710 ⇒ 00:09:10.710 Ajibade Adeleke: Yeah, so again, the first thing is to understand the business logic, which is the business requirements. So, again, after understanding the business, right, what is required to be built for the analytics side of things, so the next thing would then be
59 00:09:10.770 ⇒ 00:09:19.940 Ajibade Adeleke: ingesting from these sources into a what? Into a staging layer. Now, this involves, now, so in the…
60 00:09:24.450 ⇒ 00:09:30.929 Ajibade Adeleke: into a staging layer within… so it could depend. It depends on what kind of data processing. Got it.
61 00:09:31.460 ⇒ 00:09:35.660 Demilade Agboola: Again, you cut out again. The last thing I said was ingesting to a staging layer, so from.
62 00:09:35.660 ⇒ 00:09:36.240 Ajibade Adeleke: Yeah?
63 00:09:36.240 ⇒ 00:09:37.270 Demilade Agboola: I didn’t hear you again.
64 00:09:37.600 ⇒ 00:09:40.720 Ajibade Adeleke: Okay, okay, I’m so sorry about this.
65 00:09:41.060 ⇒ 00:09:44.179 Ajibade Adeleke: Again, I said, interestingly to the station earlier.
66 00:09:44.520 ⇒ 00:09:51.300 Ajibade Adeleke: then, since I would, I would really, probably want to use DLT,
67 00:09:51.550 ⇒ 00:10:06.980 Ajibade Adeleke: data processing techniques. So in this case, since from the transaction system, ingesting into the staging layer, then within the staging layer, within the warehouse, right, I could see DB2 could basically transform it, so this would have
68 00:10:07.120 ⇒ 00:10:21.130 Ajibade Adeleke: We could transform within 3 layers, paging, intermediate amount. So for this paging, this would just be connecting to the source system, and probably some light transformation, like, let’s say the name belongs to Camel faces,
69 00:10:21.150 ⇒ 00:10:39.820 Ajibade Adeleke: and work source, and standardizing data types. Now, the next thing we need at all this level will then be joined. So this is where, you know, even based on the business logic, joint stables together, perhaps a feature engineering that is, so, added colours, a word source.
70 00:10:39.820 ⇒ 00:10:46.719 Ajibade Adeleke: Then it serves into the world, the aggregatedly and into the data world to serve the last parliament.
71 00:10:47.660 ⇒ 00:10:48.190 Demilade Agboola: Andrew?
72 00:10:48.720 ⇒ 00:10:49.570 Demilade Agboola: Alright.
73 00:10:49.840 ⇒ 00:10:53.739 Demilade Agboola: Couple more questions based off that.
74 00:10:54.240 ⇒ 00:10:58.709 Demilade Agboola: In terms of data modeling and the math models that you’re trying to create.
75 00:10:59.210 ⇒ 00:11:03.469 Demilade Agboola: Are you… when you think of the math models you’re creating.
76 00:11:04.580 ⇒ 00:11:09.060 Demilade Agboola: Do you think of a normalized schema, or do you think of a star schema?
77 00:11:09.620 ⇒ 00:11:18.170 Demilade Agboola: And when do you utilize either? Like, they both have their strong, you know, advantages and disadvantages, so when would you want to use more of a…
78 00:11:18.540 ⇒ 00:11:21.529 Demilade Agboola: Normalized schema, or when would you want to use more of a star schema?
79 00:11:22.110 ⇒ 00:11:35.580 Ajibade Adeleke: Okay, so basically, again, you know, in the math basically has to be a little, like you said, star schema and, you know, a license form. Now, for the star schema, basically, I want to use the star schema if…
80 00:11:35.580 ⇒ 00:11:49.269 Ajibade Adeleke: we have, like, a data analyst, right? So, this is basically me preparing the old data architecture for the data analyst. So basically, this will be starting at this first dimension here.
81 00:11:49.390 ⇒ 00:11:57.779 Ajibade Adeleke: parts table having to transform, clean, and transactional process. Why the dimension table basically describe the properties of the parts table.
82 00:11:57.830 ⇒ 00:12:11.599 Ajibade Adeleke: And then, that would be the case where, probably there’s a data analyst, which will pick from the data mat. Now, for the, if I’m to choose, you know, my life form… so, you know, my life form is basically where, again, I’m serving as a data
83 00:12:12.430 ⇒ 00:12:13.970 Ajibade Adeleke: President Thailand.
84 00:12:14.950 ⇒ 00:12:26.960 Ajibade Adeleke: Right? So basically, so this is basically where I build a aggregated transformational layer within that data mat. So, let’s say you wanted to see,
85 00:12:27.240 ⇒ 00:12:35.210 Ajibade Adeleke: let’s say percentage, they wanted to see some trends, they want to see… so basically some aggregated areas regarding it. So that is when I would choose the normalized.
86 00:12:36.970 ⇒ 00:12:38.350 Demilade Agboola: Okay,
87 00:12:42.690 ⇒ 00:12:45.819 Demilade Agboola: I think a follow-up question to that will be…
88 00:12:50.540 ⇒ 00:12:53.560 Demilade Agboola: If you have a normalized schema.
89 00:12:55.990 ⇒ 00:12:59.690 Demilade Agboola: Okay, so when would you balance both? So, what I mean by that is this.
90 00:13:00.710 ⇒ 00:13:04.309 Demilade Agboola: If you have… It’s asking, man.
91 00:13:05.100 ⇒ 00:13:11.240 Demilade Agboola: for you to be able to build that report with a star schema, yes, you have to do some joins and some transformation in your BI tool.
92 00:13:12.310 ⇒ 00:13:14.300 Demilade Agboola: But, potentially.
93 00:13:14.480 ⇒ 00:13:18.699 Demilade Agboola: You can also do a mix. You can have some star schemas, and then you do your joins.
94 00:13:18.840 ⇒ 00:13:25.749 Demilade Agboola: in your BI tool, or you can have a normalized schema where you kind of just feed it in, and you can just select the columns you need to see.
95 00:13:27.730 ⇒ 00:13:40.080 Demilade Agboola: There are certain times that the level of granularity of your data does not allow you to just use a reported schema, especially if you are going to add a filter that changes granularity levels.
96 00:13:40.310 ⇒ 00:13:50.379 Demilade Agboola: My question to you is… How do you go about… deciding… what dashboards?
97 00:13:50.970 ⇒ 00:13:55.470 Demilade Agboola: And what scenarios you would want to lean on a normalized schema versus a star schema.
98 00:13:57.830 ⇒ 00:13:58.220 Demilade Agboola: Technical.
99 00:14:00.070 ⇒ 00:14:07.870 Ajibade Adeleke: Okay, so taking off cylinders. So, in this case, I would say, again, seeing as, you know,
100 00:14:08.240 ⇒ 00:14:18.759 Ajibade Adeleke: all of, like, the analytics dashboards that you can probably be able to see what you say, so high anxiety, so I, Aussie School.
101 00:14:18.990 ⇒ 00:14:38.190 Ajibade Adeleke: what is this thing called, let’s say, some Jewish food filters, right, where, you know, within, you know, within it, you can basically enter within, like, just, like, more of, like, a payroll, right? Now, in this case, also, there would also be, let’s say, want to make use of some strong,
102 00:14:38.340 ⇒ 00:14:54.879 Ajibade Adeleke: slicer capabilities. So, in the sense that what… if, the filter section side of things, let’s say you go to, you know, some of the categories available, which is applied within the slicer, which can basically filter the whole dashboard, right? So, in this case, I would lean more of starting
103 00:14:55.080 ⇒ 00:15:06.409 Ajibade Adeleke: Because basically, the star schema gives us, the ability to, you know, in a sense… so the difference… why… why are we defense star schema in that… in this case, I don’t understand what?
104 00:15:06.900 ⇒ 00:15:23.929 Ajibade Adeleke: For start schema, the dimensional tables are medically defiled, let’s say, in a, I would say, in law, right? So in this case, whether, those categorical variables are within the practice table or not, it depends.
105 00:15:24.260 ⇒ 00:15:33.289 Ajibade Adeleke: So in this way, within the dashboard side of things, even actually including that particular dimension, you’re able to see, okay, now.
106 00:15:33.400 ⇒ 00:15:41.959 Ajibade Adeleke: Whether it’s impacted, the transaction is within factable or not, if you select it, you’re saying that the value itself.
107 00:15:42.260 ⇒ 00:15:48.169 Ajibade Adeleke: Right, so that’s the way now we can also, start…
108 00:15:49.730 ⇒ 00:16:04.700 Ajibade Adeleke: Not just based on what the data we have. But if it is just probably a quick one, not a very detailed report, just based on what we really have, then I would lean on the normalization perspective.
109 00:16:05.150 ⇒ 00:16:06.010 Demilade Agboola: Okay.
110 00:16:06.570 ⇒ 00:16:07.990 Demilade Agboola: Alright.
111 00:16:08.740 ⇒ 00:16:09.209 Ajibade Adeleke: That’s a lot.
112 00:16:09.210 ⇒ 00:16:16.600 Demilade Agboola: Let’s say we’ve built out that entire… Our entire, dbt pipeline.
113 00:16:17.790 ⇒ 00:16:25.469 Demilade Agboola: And we realized that we have one of our models that is quite slow, right? So we see that it’s, like, it takes…
114 00:16:25.620 ⇒ 00:16:27.850 Demilade Agboola: Two hours to run just one model.
115 00:16:28.320 ⇒ 00:16:38.369 Demilade Agboola: So I’m just… this is an extreme case, so let’s say we have, what, 500 million rows, 1 billion rows, whatever number. It’s just a very large model, lots of data in there.
116 00:16:38.990 ⇒ 00:16:43.240 Demilade Agboola: And… We come to you and say, hey, how do we…
117 00:16:43.870 ⇒ 00:16:53.250 Demilade Agboola: optimize this query. So what will you be looking for within that query, and how… what processes are you going to just be thinking about?
118 00:16:53.500 ⇒ 00:16:56.080 Demilade Agboola: To make that model run faster.
119 00:16:57.440 ⇒ 00:17:03.949 Ajibade Adeleke: Okay, one of the first… so, since this is within a DBT, so one of the first steps will be…
120 00:17:04.819 ⇒ 00:17:07.800 Ajibade Adeleke: First check of re-optimization.
121 00:17:09.200 ⇒ 00:17:12.920 Demilade Agboola: Sorry, can you repeat yourself? Again, you went… your voice went out, and…
122 00:17:13.089 ⇒ 00:17:14.269 Ajibade Adeleke: Okay, bueno.
123 00:17:14.880 ⇒ 00:17:19.499 Demilade Agboola: Yeah, it’s a bit better. Could also be your headphones, I’m not sure if they’re properly set.
124 00:17:21.060 ⇒ 00:17:26.820 Ajibade Adeleke: Yeah, I think, I think it’s basically quite useful.
125 00:17:27.020 ⇒ 00:17:27.990 Demilade Agboola: Oh, okay.
126 00:17:29.220 ⇒ 00:17:32.099 Ajibade Adeleke: So, again, please confirm with me, please.
127 00:17:32.280 ⇒ 00:17:33.389 Demilade Agboola: A second here, you know.
128 00:17:33.880 ⇒ 00:17:48.890 Ajibade Adeleke: Okay, so again, so the first thing I want to do, so I basically want to implement, two strategies, right? The first primary strategy is, again, if this query is slow, so most likely they are basically using April Refresh.
129 00:17:49.180 ⇒ 00:18:02.409 Ajibade Adeleke: Right? So that is for each time the BBT is going on, it’s been fully refreshing, like, from beginning to the end, right? And the second thing is… now, I’ll just get back to the director logo.
130 00:18:03.070 ⇒ 00:18:04.860 Ajibade Adeleke: We’re going out again. It’s trusted.
131 00:18:05.280 ⇒ 00:18:11.470 Demilade Agboola: I didn’t hear the second one. I heard everything until, like, refresh, dbt loading every time. Okay, after that, I didn’t hear the…
132 00:18:12.150 ⇒ 00:18:27.530 Ajibade Adeleke: Alright, so, I mean, the second thing, right, so just, like, the primary thing will just be the incremental load, so implement and incremental load. I think most likely why the code is being really slow would be it’s using a full refresh capability.
133 00:18:27.820 ⇒ 00:18:34.259 Ajibade Adeleke: Now, for… now, the second one, right, which you just do… which is just taking an extra step.
134 00:18:34.280 ⇒ 00:18:53.290 Ajibade Adeleke: Which is basically to optimize for you to check, if some joints, to know that what… to know that what some second… some of the columns that are not being used, are not being selected, right, in this sense. Now, back to the incremental injection. So basically, you go to implement an incremental injection that is a window.
135 00:18:53.290 ⇒ 00:18:58.019 Ajibade Adeleke: So in the sense that what, since our divine has not grown large.
136 00:18:58.140 ⇒ 00:19:02.819 Ajibade Adeleke: We’d want to basically do impact authorization that is based on the range of.
137 00:19:02.820 ⇒ 00:19:04.649 Demilade Agboola: So let’s give this way to it.
138 00:19:04.650 ⇒ 00:19:15.759 Ajibade Adeleke: is… has actually been loaded, but to all new data, you’re basically loading only tomorrow data, right? In this sense, our solar support. So that will greatly optimize
139 00:19:16.160 ⇒ 00:19:17.440 Ajibade Adeleke: This is local.
140 00:19:18.350 ⇒ 00:19:19.090 Demilade Agboola: Okay.
141 00:19:20.570 ⇒ 00:19:21.950 Demilade Agboola: Alright,
142 00:19:23.400 ⇒ 00:19:30.880 Demilade Agboola: Couple more questions I’m thinking about. One is, what would you say your technical stack is? So, how many tools have you worked with?
143 00:19:31.260 ⇒ 00:19:34.950 Demilade Agboola: again, don’t… there’s no need to overemphasize, like, I…
144 00:19:35.160 ⇒ 00:19:46.050 Demilade Agboola: I, personally, I value depth, rather than, oh, I just walked one time with one tool, because if you have depth in one tool, usually it’s very easy for you to transfer from one skill to another.
145 00:19:46.940 ⇒ 00:19:50.889 Demilade Agboola: Things that you’re comfortable and confident working in, or working with.
146 00:19:51.700 ⇒ 00:19:57.380 Ajibade Adeleke: Okay, so for the technical skills I’m confident I’m working with will be Python, Okay. SQL…
147 00:19:58.670 ⇒ 00:20:14.459 Ajibade Adeleke: So, I’ve… I’ve also used Pipetran. So, you know, Pipetran is not… it’s just an EV connected to a source and then to a destination. I’ve worked with Pipetran, and then dbt as well. I’ve also worked with DBT. So, on the dashboard side of things, I’m mostly commissioned to Power BI.
148 00:20:14.840 ⇒ 00:20:21.370 Ajibade Adeleke: I’ve used that blue, Luca, before for my proficiencies within 5 years.
149 00:20:22.110 ⇒ 00:20:27.990 Demilade Agboola: Power BI, okay, cool. So let’s say you’re on a client.
150 00:20:29.020 ⇒ 00:20:33.790 Demilade Agboola: And you’ve built out your… this entire platform that we’ve talked about just now.
151 00:20:34.050 ⇒ 00:20:36.610 Demilade Agboola: How do you ensure that the data
152 00:20:36.790 ⇒ 00:20:39.349 Demilade Agboola: That they get is of the highest quality.
153 00:20:40.810 ⇒ 00:20:44.290 Ajibade Adeleke: So that would be introducing something called DCAP monitoring.
154 00:20:44.940 ⇒ 00:20:46.320 Ajibade Adeleke: and also mobility.
155 00:20:46.420 ⇒ 00:20:51.440 Ajibade Adeleke: Right? So, data coverage is basically checking the loans on loans.
156 00:20:51.660 ⇒ 00:21:06.610 Ajibade Adeleke: For data availability, we are checking your loans or your money, right? So for data quality, so this is where data quality will be put into place. Let’s say for finite case, BBT has this functionality called, this genetic test or genetic test, where
157 00:21:08.390 ⇒ 00:21:16.650 Ajibade Adeleke: for the family keys, then for the, sorry, check for the unique values within the family key, and then,
158 00:21:16.650 ⇒ 00:21:34.769 Ajibade Adeleke: Lots knows, also being locked laws, and then I would also check for accepted values. Accepted values is just basically where, you know, if you find, okay, these are the values we want to see with every school, because, like, most popular apply to, let’s say, status school, then the payment tracker, where would happen, complete, and so on.
159 00:21:34.770 ⇒ 00:21:44.919 Ajibade Adeleke: And again, this is just basically making sure that what the data quality is well put in place. And also, I’ll also put in source fresh sharing.
160 00:21:44.980 ⇒ 00:21:48.619 Ajibade Adeleke: So, I didn’t come across an issue while working
161 00:21:48.890 ⇒ 00:21:59.760 Ajibade Adeleke: when I was basically doing a modeling for a payment platform, right, that I mentioned earlier. So, based on how, you know, so basically, the platform is basically
162 00:22:00.100 ⇒ 00:22:01.570 Ajibade Adeleke: based on example.
163 00:22:04.450 ⇒ 00:22:11.170 Demilade Agboola: You went out for a brief second there, I didn’t hear what you said. Basically, the payment platform had the thing where, and then I didn’t hear anything.
164 00:22:11.170 ⇒ 00:22:26.679 Ajibade Adeleke: Yeah, so, so after I finished building, all of, like, the analytics are using B2P for transformation, and then, on top, the sense that what… they were about for 36 hours, they were still data.
165 00:22:28.110 ⇒ 00:22:44.190 Ajibade Adeleke: Right? So, in this sense, what happened was that what, based on the change of things authentication, right? So, for 5chan, stopped that connection, right? So, basically, the activity was not just stopping.
166 00:22:45.080 ⇒ 00:22:48.189 Ajibade Adeleke: So, in that process, that’s when I went to dip into
167 00:22:48.690 ⇒ 00:23:01.439 Ajibade Adeleke: This test, when was the last time you had explained? It’s supposed to be stable?
168 00:23:02.070 ⇒ 00:23:05.030 Ajibade Adeleke: updated.
169 00:23:05.030 ⇒ 00:23:05.920 Demilade Agboola: Okay.
170 00:23:06.520 ⇒ 00:23:12.399 Demilade Agboola: Alright, quick question. Are you comfortable using dbt Core or dbt Cloud?
171 00:23:12.750 ⇒ 00:23:13.990 Demilade Agboola: Specifically.
172 00:23:14.770 ⇒ 00:23:19.069 Ajibade Adeleke: Mmm… I’m confidently using dbt Pro.
173 00:23:19.310 ⇒ 00:23:22.920 Ajibade Adeleke: But I’ve also explored DBT Cloud for lots of index.
174 00:23:23.100 ⇒ 00:23:31.509 Ajibade Adeleke: I understand that, you know, DBT basically just makes you distress in which way you have to set up for DBT code.
175 00:23:31.630 ⇒ 00:23:38.410 Ajibade Adeleke: Right? And you’ll also be able to, do CIC, right? Something that’s account across.
176 00:23:40.070 ⇒ 00:23:46.010 Demilade Agboola: Okay, let’s see… I…
177 00:23:50.240 ⇒ 00:24:04.110 Demilade Agboola: how… if you were to explore a data source that you have never really used before, like, I don’t know what data sources you’ve modeled, but let’s just say you’ve not explored a certain data source, let’s say QuickBooks, or, you know, whatever data source, really.
178 00:24:04.540 ⇒ 00:24:14.549 Demilade Agboola: So now, you want to QA the data. How do you get your data QA-ready? Like, how do you go about your process to ensure that, okay, what you’re modeling
179 00:24:15.020 ⇒ 00:24:17.020 Demilade Agboola: is close to…
180 00:24:17.480 ⇒ 00:24:22.720 Demilade Agboola: what you need to, like, what’s your process to be able to discover… go from, this is all the raw data.
181 00:24:23.280 ⇒ 00:24:39.430 Demilade Agboola: I have never… I’ve never modeled this data source before. How do you go from that to numbers that you can QA internally before you say, hey, I would show this to the client? What would your process be if you were working, like, within our team, how would that look like?
182 00:24:41.250 ⇒ 00:24:43.769 Ajibade Adeleke: Hmm… let’s see…
183 00:24:46.950 ⇒ 00:24:50.660 Ajibade Adeleke: So in this sense,
184 00:24:51.280 ⇒ 00:25:01.849 Ajibade Adeleke: So, again, so, I mean, I mean, this is… this is what every… that is a very good question, the sense that what… when you are dealing with a new data source, how would you know what data is that?
185 00:25:02.020 ⇒ 00:25:10.589 Ajibade Adeleke: How’s it affecting the development?
186 00:25:11.040 ⇒ 00:25:16.809 Ajibade Adeleke: So, in this sense, so, again, this would just be for Canadian people.
187 00:25:17.280 ⇒ 00:25:22.930 Ajibade Adeleke: So this is just, in a general sense, again, connects to…
188 00:25:23.110 ⇒ 00:25:35.829 Ajibade Adeleke: data source based on any tool I’m using, and then, connected to the destination source. So, within that, the end of this is, and we just do, again, this will just be based on the basis.
189 00:25:36.230 ⇒ 00:25:50.890 Ajibade Adeleke: So this is where I’ll be able to rely on data cost to basically grabs how the data system works. I, for one, believe I want for you to basically understand, anything data.
190 00:25:50.900 ⇒ 00:26:07.209 Ajibade Adeleke: your data perspective or data platform. You need to understand the data. When you understand the data, it becomes more of, like, like, you become more, sorry to say, like, you become more… so in this case, you’ll be able to know, okay, even if, you know, there’s some data inconsistency within the data.
191 00:26:07.210 ⇒ 00:26:08.899 Demilade Agboola: You’ll be able to check in, okay.
192 00:26:09.140 ⇒ 00:26:17.290 Ajibade Adeleke: I know where this is coming from. So, in this case, I would rely very heavily on data, quality, data validity.
193 00:26:19.660 ⇒ 00:26:26.289 Demilade Agboola: Okay, alright, final question will probably be…
194 00:26:27.650 ⇒ 00:26:29.829 Demilade Agboola: If a client wants a dashboard.
195 00:26:29.950 ⇒ 00:26:32.289 Demilade Agboola: But they’re not clear about their metrics.
196 00:26:34.090 ⇒ 00:26:41.930 Demilade Agboola: And they want you to do, like, just model things in a way That is not scalable, right?
197 00:26:42.610 ⇒ 00:26:47.090 Demilade Agboola: How would you go about client management in that scenario?
198 00:26:49.490 ⇒ 00:27:03.510 Ajibade Adeleke: So, again, their… their endpoint is not clear, so I need to build back in to be scalable. Now, in this sense, what I’ve been currently doing is, this is something I’ve actually had experienced.
199 00:27:03.960 ⇒ 00:27:05.230 Ajibade Adeleke: These are late.
200 00:27:06.920 ⇒ 00:27:07.799 Demilade Agboola: Sorry, you’re cutting out again.
201 00:27:09.940 ⇒ 00:27:11.490 Ajibade Adeleke: Thank you.
202 00:27:12.440 ⇒ 00:27:13.959 Demilade Agboola: Yes, I can hear you now.
203 00:27:14.870 ⇒ 00:27:15.430 Ajibade Adeleke: But…
204 00:27:16.130 ⇒ 00:27:27.839 Ajibade Adeleke: So, as I was saying, again, this is something I’ve had experience with. Also, the local ownership involves a lot of candidates.
205 00:27:27.990 ⇒ 00:27:31.899 Ajibade Adeleke: communication with stakeholders and whatsoever. So, in this sense.
206 00:27:32.060 ⇒ 00:27:49.409 Ajibade Adeleke: So mostly, sometimes stakeholders do not even actually know what they want. So in this case, for me, building a scalable data model, would they just be, so, again, this is just me looking at the data, looking at what is essentially the attributes, and…
207 00:27:49.700 ⇒ 00:27:59.710 Ajibade Adeleke: making sure that I want to… this is, again, this is where I actually, where my advanced data managers are at, right? So, making sure that I want to take this data for the base, what I want…
208 00:27:59.710 ⇒ 00:28:10.230 Ajibade Adeleke: based on the, industry as well. So, again, it’s also depending on industry. I know motels, they have a hotel for a fintech, so our motel for, let’s say.
209 00:28:16.800 ⇒ 00:28:18.440 Ajibade Adeleke: Actually, I’ll…
210 00:28:18.440 ⇒ 00:28:24.520 Demilade Agboola: No, no. I heard you model, like, fintech and governments, e-government.
211 00:28:24.520 ⇒ 00:28:25.200 Ajibade Adeleke: Yeah.
212 00:28:25.860 ⇒ 00:28:43.470 Ajibade Adeleke: again, a lot of, research and a lot of funding, so those are two most likely, metrics that are being useful to a particular industry. So it’s heavily based on an industry case and based on self-reach that are going to be determined by scale.
213 00:28:44.300 ⇒ 00:28:45.170 Demilade Agboola: Oh, okay.
214 00:28:45.760 ⇒ 00:28:54.460 Demilade Agboola: Alright, do you have any questions about, you know, Brainforge? Any questions for me? Or anything?
215 00:28:54.460 ⇒ 00:28:59.810 Ajibade Adeleke: I think my question would be, you know, What does your B2B insight?
216 00:29:01.370 ⇒ 00:29:05.609 Demilade Agboola: So my day-to-day varies quite tremendously,
217 00:29:06.780 ⇒ 00:29:22.289 Demilade Agboola: So, basically, I work with… I currently work with 3 different clients, so right now, that’s what my day-to-day looks like. I work with 3 different clients. Don’t worry, most people that come in, especially initially, you’re just on one, and eventually two.
218 00:29:22.520 ⇒ 00:29:24.620 Demilade Agboola: I’m on 3, cause…
219 00:29:26.050 ⇒ 00:29:32.910 Demilade Agboola: I am kind of higher up, so that… they put me on a number of things, so I can, like, multitask across different clients.
220 00:29:33.160 ⇒ 00:29:45.029 Demilade Agboola: But day-to-day varies. So we have stand-ups in the morning, US time, so that would be, like, afternoon hour time, because I’m in Malta, so my time zone is right… is the same as, I believe you’re in Nigeria, right?
221 00:29:45.610 ⇒ 00:29:49.389 Demilade Agboola: So, the… our times are exactly the same right now.
222 00:29:49.760 ⇒ 00:29:57.760 Demilade Agboola: So my… I basically start off with a bunch of meetings, so we have… stand-ups.
223 00:29:58.170 ⇒ 00:30:03.879 Demilade Agboola: And then we have, like, I usually have, like, some other meetings with, like, CSO, sometimes with, like.
224 00:30:04.720 ⇒ 00:30:08.760 Demilade Agboola: Well, Tom… What time is our CEO, or some other people.
225 00:30:09.210 ⇒ 00:30:13.199 Demilade Agboola: And then eventually, I have to put aside time for work.
226 00:30:13.320 ⇒ 00:30:17.759 Demilade Agboola: So I have, like, 4 hours in my day where I sit down, I go through the different models.
227 00:30:17.930 ⇒ 00:30:21.890 Demilade Agboola: And then I modeled different things for different clients.
228 00:30:22.120 ⇒ 00:30:24.830 Demilade Agboola: Sometimes push documentation.
229 00:30:24.940 ⇒ 00:30:30.130 Demilade Agboola: About what I have modeled, so that whoever is showing it to the client can have a better
230 00:30:30.380 ⇒ 00:30:33.460 Demilade Agboola: Understanding of the logic that was embedded in the data.
231 00:30:34.110 ⇒ 00:30:43.229 Demilade Agboola: And then usually, again, towards the end of the day, I tend to have meetings as well. Again, part of the whole thing where I…
232 00:30:44.090 ⇒ 00:30:55.060 Demilade Agboola: another meeting, like, OTAM, so we have CSOs, so those are, like, client success owners, so I’m also a part of that, so I’m… I would have meetings around that as well, in the evenings.
233 00:30:55.720 ⇒ 00:31:03.320 Demilade Agboola: But yes, it’s usually around that. I’m usually either modeling data in meetings, building a dashboard.
234 00:31:03.420 ⇒ 00:31:13.849 Demilade Agboola: Or, hopping on a call to help someone on the team put off their fire. So, usually, like, on different projects, people reach out to me if certain things are going sideways.
235 00:31:14.030 ⇒ 00:31:19.519 Demilade Agboola: Sometimes the clients themselves reach out to me if they feel like things are not going the way they want it to go.
236 00:31:19.890 ⇒ 00:31:24.419 Demilade Agboola: And then I just hop in and try and help things, you know, align things, push them forward.
237 00:31:26.240 ⇒ 00:31:39.809 Ajibade Adeleke: Alright, that sounds interesting. So, another thing I would… I would want to add, based on your answer was, you said building that. So, to be honest, can you start? Like, you propose what kind of stack
238 00:31:39.970 ⇒ 00:31:49.570 Ajibade Adeleke: The client’s needs based on their business complain, or, you know, they bring it with their staff and say, hey, we need this to be,
239 00:31:50.460 ⇒ 00:31:58.500 Ajibade Adeleke: Built, right, on an analytics scale, and also, when you say client, you just build a one-stop solution?
240 00:31:59.020 ⇒ 00:32:05.540 Ajibade Adeleke: Because I know, Brave was constantly, more, like, they see themselves from, like, the partner, right? I think.
241 00:32:05.730 ⇒ 00:32:13.890 Ajibade Adeleke: So just to fix that. It’s like having you just being in data for data.
242 00:32:14.310 ⇒ 00:32:16.000 Ajibade Adeleke: See you.
243 00:32:18.110 ⇒ 00:32:21.730 Demilade Agboola: So the first answer to… the answer to your first question is it depends.
244 00:32:21.880 ⇒ 00:32:24.080 Demilade Agboola: For some people, they come in.
245 00:32:24.400 ⇒ 00:32:30.719 Demilade Agboola: they already have infrastructure, so we just need to make it work. Like, if you have a top… if you have a contract with Tableau for the next 2 years.
246 00:32:31.410 ⇒ 00:32:39.140 Demilade Agboola: we can’t ask you to necessarily cut that, because we feel there’s a different BI tool, for instance, that will be better for you. We will make Tableau work.
247 00:32:40.760 ⇒ 00:32:50.189 Demilade Agboola: But in some cases, especially when the person hasn’t, like, formulated, or the client hasn’t formulated everything, we can make recommendations. We can get an idea of what’s your budget.
248 00:32:50.390 ⇒ 00:33:03.800 Demilade Agboola: who are the main users, how would they utilize it, what do they need to see, and we can start using all that information to say, hey, if you’re going to build out this sort of dashboard, it’s a bit complex to build that in Tableau, for instance. This tool might be easier.
249 00:33:03.960 ⇒ 00:33:08.979 Demilade Agboola: For instance, we’ve been using a tool a lot called Omni, you can check it out.
250 00:33:09.650 ⇒ 00:33:16.820 Demilade Agboola: Omni basically is pretty good for building dashboards. It’s pretty solid, you can build spreadsheets, like, spreadsheet formulas as well.
251 00:33:16.930 ⇒ 00:33:34.369 Demilade Agboola: Also, you can also do, like, they have a lot of, like, AI features, so you can build out things with AI quickly, but the advantage of that is if you build out good data sets, you can also put AI context, and so business stakeholders who don’t necessarily
252 00:33:34.670 ⇒ 00:33:47.169 Demilade Agboola: who can actually normally build dashboards or do anything there? They can go to those datasets and ask questions directly and say, hey, even if there’s no dashboard for it, what are my sales been like over the last 14.
253 00:33:47.170 ⇒ 00:33:50.010 Ajibade Adeleke: You know, to just bring me, like, coffee.
254 00:33:50.180 ⇒ 00:33:51.220 Ajibade Adeleke: Yeah. Awesome.
255 00:33:51.470 ⇒ 00:34:08.570 Demilade Agboola: We can make recommendations like that based off of that, but, like, again, if you already have a contract, and you don’t want to, because it would cost you money to break that contract, we can say, okay, cool, we will make… do with what you have, and we will not recommend anything for now, but long-term, this is our recommendation on what you should be using.
256 00:34:08.850 ⇒ 00:34:12.160 Demilade Agboola: So there’s that. The second question was about, like.
257 00:34:12.290 ⇒ 00:34:16.310 Demilade Agboola: Partnership, and yeah, we try our best to…
258 00:34:17.040 ⇒ 00:34:19.760 Demilade Agboola: give recommendations. We remember that, like…
259 00:34:20.949 ⇒ 00:34:23.180 Demilade Agboola: While we are their data team.
260 00:34:23.340 ⇒ 00:34:25.729 Demilade Agboola: We try to see ourselves more as, like.
261 00:34:25.860 ⇒ 00:34:35.540 Demilade Agboola: how can we direct you and help you make as much money and as much progress in your business as possible? Because that’s what, effectively, we’re being paid for.
262 00:34:35.580 ⇒ 00:34:47.079 Demilade Agboola: It’s not just to show you, okay, here’s a dashboard of everything. No, it’s like, how can we empower you, and how can we be sure that the decisions you’re making are in the right direction for your company?
263 00:34:47.219 ⇒ 00:34:52.380 Demilade Agboola: And the best way we find ourselves doing that is by giving suggestions.
264 00:34:52.610 ⇒ 00:35:04.819 Demilade Agboola: And we like it, like, we want to become, like, more of partners, so we like people on our team who suggest things, who think long-term, who are like, hey, how about we do this analysis?
265 00:35:04.940 ⇒ 00:35:10.350 Demilade Agboola: oh, you have this data, I’ve seen it used this way, how about we try it for you?
266 00:35:10.580 ⇒ 00:35:20.539 Demilade Agboola: if there are issues that the client is having, how can we make that go away so that you’re always empowered to do what you need to do? So, things like that. We try our best to not just be…
267 00:35:20.760 ⇒ 00:35:27.610 Demilade Agboola: Data seems like, yes, there are people, but actually people who are producing value.
268 00:35:27.610 ⇒ 00:35:30.949 Ajibade Adeleke: Like, all of, like, let’s part of BTX and BRC.
269 00:35:31.290 ⇒ 00:35:33.210 Demilade Agboola: Exactly, exactly.
270 00:35:35.010 ⇒ 00:35:35.900 Demilade Agboola: Okay.
271 00:35:36.290 ⇒ 00:35:37.060 Demilade Agboola: All right.
272 00:35:37.060 ⇒ 00:35:38.510 Ajibade Adeleke: Thank you for that.
273 00:35:39.410 ⇒ 00:35:40.780 Ajibade Adeleke: No, no, that’s… that’s…
274 00:35:40.970 ⇒ 00:35:50.699 Demilade Agboola: Sounds good. Thank you so much for this. I will write up my notes, and I’ll send it to, Kayla. Kayla is our head of recruitment, and she’ll be in touch with you.
275 00:35:52.420 ⇒ 00:35:56.299 Demilade Agboola: Alright, thank you very much. Okay, yeah. Take care.
276 00:35:57.780 ⇒ 00:35:58.470 Ajibade Adeleke: holds it.