Meeting Title: Brainforge Interview w- Awaish Date: 2026-03-24 Meeting participants: Awaish Kumar, Hussein Diab
WEBVTT
1 00:00:05.480 ⇒ 00:00:06.300 Awaish Kumar: Hi.
2 00:00:07.520 ⇒ 00:00:08.720 Hussein Diab: Hello, hello!
3 00:00:09.880 ⇒ 00:00:11.370 Awaish Kumar: Hi, San, how you doing?
4 00:00:11.650 ⇒ 00:00:13.500 Hussein Diab: I’m doing well. How are you?
5 00:00:13.930 ⇒ 00:00:16.589 Awaish Kumar: I’m good as well. Where are you located?
6 00:00:17.270 ⇒ 00:00:18.520 Hussein Diab: I’m in Dallas.
7 00:00:19.820 ⇒ 00:00:21.530 Awaish Kumar: Okay, yeah.
8 00:00:21.530 ⇒ 00:00:22.239 Hussein Diab: Where are you?
9 00:00:22.530 ⇒ 00:00:27.900 Awaish Kumar: 30-minute session. I mean, it’s, like, just to know more about you,
10 00:00:28.660 ⇒ 00:00:33.720 Awaish Kumar: And, like, I walk over the projects we have done.
11 00:00:33.900 ⇒ 00:00:43.649 Awaish Kumar: And then, just, if you have any questions regarding brain folds or what we do, I will be there to answer that. Yeah, that’s basically it.
12 00:00:44.300 ⇒ 00:00:46.569 Hussein Diab: Definitely. Where are you located, Avish?
13 00:00:47.100 ⇒ 00:00:48.530 Awaish Kumar: I’m in UAE right now.
14 00:00:50.550 ⇒ 00:00:51.709 Hussein Diab: In UAE?
15 00:00:51.850 ⇒ 00:00:52.390 Awaish Kumar: Yep.
16 00:00:52.930 ⇒ 00:00:56.069 Hussein Diab: Oh, okay, alright. Do you speak Arabic there?
17 00:00:56.710 ⇒ 00:00:57.320 Awaish Kumar: Oops.
18 00:00:57.670 ⇒ 00:00:58.880 Hussein Diab: Do you speak Arabic?
19 00:00:59.860 ⇒ 00:01:00.849 Awaish Kumar: Oh, no, no.
20 00:01:01.070 ⇒ 00:01:05.019 Awaish Kumar: I don’t know. I’m trying to learn. I just came by.
21 00:01:06.650 ⇒ 00:01:08.039 Hussein Diab: Awesome, man. That’s awesome.
22 00:01:09.280 ⇒ 00:01:12.679 Awaish Kumar: Yeah, we’ll be able to speak maybe after a year or something.
23 00:01:12.940 ⇒ 00:01:14.210 Hussein Diab: Yep.
24 00:01:14.580 ⇒ 00:01:15.920 Hussein Diab: It’s a hard language.
25 00:01:17.690 ⇒ 00:01:24.600 Awaish Kumar: Okay, so let’s dive into it, like, can you, like, introduce yourself?
26 00:01:25.120 ⇒ 00:01:28.989 Hussein Diab: Sure, for sure, yeah. Do you mind if I share my screen?
27 00:01:30.540 ⇒ 00:01:31.140 Awaish Kumar: Yep.
28 00:01:31.680 ⇒ 00:01:32.630 Awaish Kumar: Alright.
29 00:01:34.530 ⇒ 00:01:36.640 Hussein Diab: There you go. Can you see it?
30 00:01:39.530 ⇒ 00:01:40.260 Hussein Diab: Alright.
31 00:01:40.830 ⇒ 00:01:47.950 Hussein Diab: So, I’m Hussein. I’m originally from Lebanon. I’ve been in Dallas for the past 20 years or so.
32 00:01:48.320 ⇒ 00:01:59.350 Hussein Diab: My background is, started with automotive to nursing, and then I, joined, UT Southwestern.
33 00:01:59.640 ⇒ 00:02:11.529 Hussein Diab: I published some papers there about preclinical trials, and then I joined Abbott Laboratory. I was a reliability manager there for 10 years or so.
34 00:02:11.710 ⇒ 00:02:17.819 Hussein Diab: And the past couple of years, I joined a consulting company, consulting with multiple clients.
35 00:02:18.270 ⇒ 00:02:27.979 Hussein Diab: This is my family, I’m married, I have two kids, I love soccer, I play, coach, and watch soccer, big AC Milan fan.
36 00:02:28.120 ⇒ 00:02:36.800 Hussein Diab: I play video games, and I design video games. I love coffee, love jiu-jitsu, and I love the beach and the mountains.
37 00:02:36.910 ⇒ 00:02:42.579 Hussein Diab: These are some of my, BMW Cafe Racer motorcycles that I’ve built.
38 00:02:43.400 ⇒ 00:02:44.520 Hussein Diab: Yeah.
39 00:02:44.670 ⇒ 00:02:46.260 Hussein Diab: Do you have any questions for me?
40 00:02:47.290 ⇒ 00:02:49.549 Awaish Kumar: No, all great.
41 00:02:50.710 ⇒ 00:02:58.080 Awaish Kumar: Seems like, you know, you have a lot of, activities that you participate in.
42 00:02:58.580 ⇒ 00:02:59.000 Hussein Diab: Yeah.
43 00:03:01.010 ⇒ 00:03:02.200 Awaish Kumar: Great to see that.
44 00:03:02.990 ⇒ 00:03:03.390 Hussein Diab: Awesome.
45 00:03:03.390 ⇒ 00:03:04.070 Awaish Kumar: Fair enough.
46 00:03:04.280 ⇒ 00:03:19.429 Hussein Diab: Alright, let me, move forward. So, the past 2 years, I’ve joined, Analytic Vision, and, this page, they created it for me as, like, a pre-sale for, like, with a new client or so.
47 00:03:19.700 ⇒ 00:03:32.199 Hussein Diab: So I’m a data engineering and analytics consultant. I’ve spent the last several years solving messy business problems across cloud data platform.
48 00:03:32.340 ⇒ 00:03:36.000 Hussein Diab: Dashboards, data modernization effort,
49 00:03:36.150 ⇒ 00:03:42.329 Hussein Diab: A lot of my work has sat at the intersection of business needs and technical delivery.
50 00:03:42.450 ⇒ 00:03:50.650 Hussein Diab: So, which is… one of the reasons why this role stands out to me. I enjoy taking ambiguous,
51 00:03:50.890 ⇒ 00:03:55.150 Hussein Diab: Problem and turning it into… Simple.
52 00:03:55.410 ⇒ 00:03:59.200 Hussein Diab: Solution, something the client can actually trust and use.
53 00:04:01.120 ⇒ 00:04:06.990 Hussein Diab: I’ve used many… Tools… God.
54 00:04:07.480 ⇒ 00:04:15.779 Hussein Diab: Snowflake, dbt, Databricks… You… you call it… Yeah, in a nutshell.
55 00:04:16.940 ⇒ 00:04:23.470 Awaish Kumar: Yeah, sounds great. Maybe if you can walk me through any of the… any… one of your recent projects that you did?
56 00:04:28.600 ⇒ 00:04:31.449 Awaish Kumar: Just deep dive into that project, and…
57 00:04:32.010 ⇒ 00:04:36.840 Awaish Kumar: Like, break it down into chunks, and maybe we can walk through it.
58 00:04:37.760 ⇒ 00:04:43.830 Hussein Diab: Sure, so one of the clients that I served recently is Kimberly Clark.
59 00:04:44.310 ⇒ 00:04:49.320 Hussein Diab: So, I was supporting their project for post-Event analyzer.
60 00:04:49.670 ⇒ 00:05:00.579 Hussein Diab: And this involved data movement and reporting across Snowflake, Databricks, SQL Server, GBT, ADF, and Power BI.
61 00:05:00.740 ⇒ 00:05:03.490 Hussein Diab: So the prob… the, the, the,
62 00:05:04.070 ⇒ 00:05:08.939 Hussein Diab: The problem was they were using a third-party
63 00:05:09.010 ⇒ 00:05:24.280 Hussein Diab: tool from PwC, and they were spending around $2 million on it yearly, and where I stepped in is to help recreate this tool in-house using the same data sources.
64 00:05:24.590 ⇒ 00:05:39.659 Hussein Diab: So, my job was not just to fix individual issues, I needed to help create more dependable architecture and operating pattern, so the team could stop, you know, firefighting and move toward,
65 00:05:39.880 ⇒ 00:05:41.919 Hussein Diab: released readiness.
66 00:05:42.040 ⇒ 00:05:57.080 Hussein Diab: So, I focused on the points where the system and the teams were drifting apart. So, environment-aware database logic, metadata generation, primarily key handling and orchestration.
67 00:05:57.500 ⇒ 00:06:01.310 Hussein Diab: I worked through dbt models and macro pattern.
68 00:06:01.410 ⇒ 00:06:09.799 Hussein Diab: Pipeline behavior, deployment concerns, so the team could standardize how tables were identified, loaded, validated.
69 00:06:10.190 ⇒ 00:06:15.139 Hussein Diab: And, we had 3 environments, dev, QA, and prod.
70 00:06:15.370 ⇒ 00:06:16.520 Hussein Diab: So…
71 00:06:16.680 ⇒ 00:06:27.609 Hussein Diab: I kept the communication tight with the internal and the client stakeholder team. Any technical issues, any risk, any item.
72 00:06:28.470 ⇒ 00:06:40.009 Hussein Diab: make it clarity before, create problems. The outcome was more, controlled path towards stability and go-live readiness, and
73 00:06:40.350 ⇒ 00:06:53.269 Hussein Diab: instead of treating each failure as, like, separate prized, I wrote a clearer structure of how the platform should behave across environments, and the end result was,
74 00:06:53.380 ⇒ 00:06:59.839 Hussein Diab: We were able to build this tool in-house, and we saved the client $2 million a year.
75 00:07:00.030 ⇒ 00:07:06.139 Hussein Diab: And because we built it ourself, we have the freedom to add features and add…
76 00:07:07.070 ⇒ 00:07:08.959 Hussein Diab: Whatever we want to it, you know?
77 00:07:11.760 ⇒ 00:07:16.929 Awaish Kumar: And, like, you build that tool using… platform using what, like, what rules?
78 00:07:17.970 ⇒ 00:07:18.730 Awaish Kumar: Nothing.
79 00:07:19.740 ⇒ 00:07:21.420 Hussein Diab: I’m sorry, can you say that again?
80 00:07:21.830 ⇒ 00:07:27.449 Awaish Kumar: Yeah, I mean, like, what tech stake that you used for that platform?
81 00:07:27.790 ⇒ 00:07:43.869 Hussein Diab: So, we were on Azure, Snowflake, Databricks, SQL Server, and for visualization, we were using a web app, and, like, a custom-made web app, and, Power BI for tracing.
82 00:07:44.070 ⇒ 00:07:52.620 Hussein Diab: I use dbt Core inside Databricks, so we were saving a lot of money on dbt Cloud.
83 00:07:52.930 ⇒ 00:07:59.570 Hussein Diab: And, the whole thing was orchestrated using Azure Data Factory, ADF.
84 00:08:02.950 ⇒ 00:08:11.210 Awaish Kumar: Okay, and then you were already using Azure Data Factory and the Databricks that you mentioned, so why…
85 00:08:11.340 ⇒ 00:08:13.810 Awaish Kumar: Snowflake was also used.
86 00:08:15.000 ⇒ 00:08:15.720 Awaish Kumar: Nope.
87 00:08:17.120 ⇒ 00:08:18.880 Hussein Diab: why Snowflake was used.
88 00:08:18.960 ⇒ 00:08:37.640 Hussein Diab: So, all of these sources originally come inside a Unity catalog inside Snowflake. So, the way we built the web app, these sources, they land in Snowflake first, so we do have the freedom to use Snowflake notebooks to create our own tables and our own views.
89 00:08:38.700 ⇒ 00:08:47.599 Hussein Diab: And we’re using dbt, within Databricks to recreate the materialized tables and do the transformation.
90 00:08:49.310 ⇒ 00:08:51.849 Hussein Diab: This way, it will run daily or weekly.
91 00:08:53.000 ⇒ 00:09:00.339 Awaish Kumar: Okay, so the… so the sources were landing the data into Snowflake, and then you were moving the data from
92 00:09:00.480 ⇒ 00:09:03.320 Awaish Kumar: Snowflake to… Databricks?
93 00:09:03.790 ⇒ 00:09:06.540 Awaish Kumar: And then running VBT on top of it?
94 00:09:06.540 ⇒ 00:09:07.420 Hussein Diab: That’s right.
95 00:09:08.490 ⇒ 00:09:14.180 Awaish Kumar: And, like, was there any… Like…
96 00:09:14.730 ⇒ 00:09:27.229 Awaish Kumar: requirement to specifically use Databricks, like, you can run dbt in Snowflake as well, right? So, yeah, I just don’t understand your choices, like, why… by Databricks? We were.
97 00:09:27.230 ⇒ 00:09:34.579 Hussein Diab: We were in a secure environment, and that was one of the requirements inside the statement of work.
98 00:09:36.190 ⇒ 00:09:45.940 Hussein Diab: Same thing with dbt, with, like… for me personally, I would rather use dbt inside VS Code, not inside Databricks, but that was one of the ways they wanted
99 00:09:48.530 ⇒ 00:09:52.460 Hussein Diab: You know, each client has separate requirements, and…
100 00:09:52.460 ⇒ 00:09:56.490 Awaish Kumar: Okay, understood. So… Since,
101 00:09:57.300 ⇒ 00:10:05.679 Awaish Kumar: like, since you all have experience with the Databricks, like, and I also see that you have experience with using,
102 00:10:05.980 ⇒ 00:10:07.619 Awaish Kumar: No, definitely.
103 00:10:13.060 ⇒ 00:10:17.389 Awaish Kumar: Sorry, a bit of noise on my side, sorry for that.
104 00:10:17.390 ⇒ 00:10:18.130 Hussein Diab: Okay.
105 00:10:18.130 ⇒ 00:10:27.929 Awaish Kumar: Yeah, but if there is a Snowflake and Databricks, and then you also have experience with GCP, I see that you have certifications in it, so…
106 00:10:28.980 ⇒ 00:10:34.159 Awaish Kumar: What do you think the pros and cons of each, it varies.
107 00:10:36.450 ⇒ 00:10:41.359 Hussein Diab: the pros and cons of, like, Azure versus GCP versus.
108 00:10:41.360 ⇒ 00:10:44.599 Awaish Kumar: versus Snowflake versus Databricks.
109 00:10:46.860 ⇒ 00:10:51.109 Hussein Diab: The pros and cons is what the client has.
110 00:10:51.350 ⇒ 00:10:55.550 Hussein Diab: So, if the client is on… Yeah, go ahead.
111 00:10:55.550 ⇒ 00:11:08.399 Awaish Kumar: If I can rephrase my question, I want to understand what you find… what differences did you find while you were using these three different platforms? BigQuery, Snowflake, and Databricks?
112 00:11:10.230 ⇒ 00:11:15.900 Hussein Diab: So… Great, let’s see.
113 00:11:16.150 ⇒ 00:11:20.940 Hussein Diab: they’re… Separate. So,
114 00:11:24.120 ⇒ 00:11:33.230 Hussein Diab: you use them together, it’s not one or, so it’s not like, hey, you have to use Databricks or Snowflake, but you can use them together.
115 00:11:33.230 ⇒ 00:11:35.930 Awaish Kumar: I just want to see if you found any differences, like.
116 00:11:36.120 ⇒ 00:11:50.510 Awaish Kumar: for example, BigQuery and Snowflake, they both can be used as data warehouses as well. So, can you differentiate, like, when we should use BigQuery, or when we should use Snowflake, or something like that?
117 00:11:51.290 ⇒ 00:12:01.989 Hussein Diab: Yeah, so if you are on a Google platform, and you are using Looker, and you are using their CICD, PubSub.
118 00:12:03.410 ⇒ 00:12:13.140 Hussein Diab: then BigQuery would make sense to get it integrated there. If you are a Microsoft shop, and you are on Power BI, and
119 00:12:13.870 ⇒ 00:12:21.429 Hussein Diab: you have some kind of premium subscription, then it would make sense to use their SQL warehouse and ADF,
120 00:12:22.010 ⇒ 00:12:23.899 Hussein Diab: If you have…
121 00:12:24.560 ⇒ 00:12:36.959 Hussein Diab: heavy use for Spark, and you need compute, and you want to create some kind of automated jobs or anything like that, then Databricks is great.
122 00:12:37.330 ⇒ 00:12:45.200 Hussein Diab: It differs from use to use cases. There’s no tool better than another, just…
123 00:12:45.690 ⇒ 00:12:49.430 Hussein Diab: What’s the project is, what the
124 00:12:49.750 ⇒ 00:12:58.969 Hussein Diab: who is the client? What are they using? But, whether it’s AWS, GCP or Azure.
125 00:12:59.080 ⇒ 00:13:05.099 Hussein Diab: They all can do the same thing. They have the same functions,
126 00:13:05.260 ⇒ 00:13:09.990 Hussein Diab: The cost might be different, depending on what you’re looking for, but…
127 00:13:09.990 ⇒ 00:13:14.410 Awaish Kumar: Yeah, like, that’s exactly what I’m… I was looking for. I know, like.
128 00:13:14.640 ⇒ 00:13:30.180 Awaish Kumar: you can use all these warehouses, BigQuery, Snowflake, doing the same thing, but, like, maybe in some use cases, it’s better to use BigQuery than Snowflake, or… or maybe how… how they’re, basically, the pricing model works, or…
129 00:13:30.300 ⇒ 00:13:34.669 Awaish Kumar: Or the… the different techniques that they employ as a system.
130 00:13:35.180 ⇒ 00:13:35.830 Hussein Diab: Yeah.
131 00:13:36.030 ⇒ 00:13:43.209 Awaish Kumar: How they differentiate, like, for example, how optimization happens in Snowflake versus BigQuery.
132 00:13:47.190 ⇒ 00:13:51.449 Awaish Kumar: for example, one of the techniques, I would just say, like, partitioning, for example.
133 00:13:51.840 ⇒ 00:13:55.130 Hussein Diab: And by clustering and partitioning and optimization, it’s…
134 00:13:55.340 ⇒ 00:14:00.530 Awaish Kumar: So, for example, partitioning. How partitioning happens in BigCerry versus in Snowflake?
135 00:14:10.840 ⇒ 00:14:14.139 Hussein Diab: how partitioning differ in BigQuery versus Snowflake?
136 00:14:17.380 ⇒ 00:14:21.520 Hussein Diab: So, with the Snowflake, you have… the,
137 00:14:22.620 ⇒ 00:14:35.050 Hussein Diab: zero copy, for example, you can leverage it. How the partition happens, it’s… it’s the same idea. Partition, it’s the same in Snowflake and BigQuery. It’s SQL-based,
138 00:14:36.990 ⇒ 00:14:37.850 Hussein Diab: Okay, so…
139 00:14:37.850 ⇒ 00:14:38.200 Awaish Kumar: Oh, God.
140 00:14:38.200 ⇒ 00:14:40.910 Hussein Diab: Snowflake will have more…
141 00:14:47.380 ⇒ 00:14:50.659 Awaish Kumar: Okay, no worries. In Snowflake, it is actually…
142 00:14:51.160 ⇒ 00:14:56.069 Awaish Kumar: Auto-partitioning, so they, they automate, like, they themselves do micro-partitioning.
143 00:14:56.300 ⇒ 00:15:05.469 Awaish Kumar: Internally, it’s part of their own system. They do optimizations, and in BigQuery, actually, they don’t do it automatically.
144 00:15:05.470 ⇒ 00:15:17.440 Awaish Kumar: a user has to define the partitioning, like, I want a partitioning for this table on this specific column, or maybe if I want to use time range partitioning, or…
145 00:15:17.600 ⇒ 00:15:28.670 Awaish Kumar: using integer range partitioning versus time unit column or something. In BigQuery, you have to do that. If you don’t do that, it won’t be…
146 00:15:28.820 ⇒ 00:15:37.259 Awaish Kumar: you don’t have partitioning, and it will just add up to the cost. And Snowflake, it just will do it as part of their system.
147 00:15:38.020 ⇒ 00:15:38.760 Hussein Diab: Okay.
148 00:15:38.760 ⇒ 00:15:47.449 Awaish Kumar: We can move on. We can move on to, more, like, questions related to DBT. I have, like.
149 00:15:48.690 ⇒ 00:15:54.680 Awaish Kumar: Can you define why we should be using dbt? Because dbt is also essentially writing SQL, right?
150 00:15:54.820 ⇒ 00:15:58.129 Awaish Kumar: So, why we actually use DVD?
151 00:15:59.700 ⇒ 00:16:08.839 Hussein Diab: For repeatable macros, or snapshot, or materialization, utilizing, leveraging Jinja.
152 00:16:09.570 ⇒ 00:16:14.859 Hussein Diab: So you don’t have to write the same thing over and over, you can just have shortcuts there.
153 00:16:15.250 ⇒ 00:16:20.400 Hussein Diab: You can also have dbt tests to test the data quality.
154 00:16:21.060 ⇒ 00:16:28.050 Hussein Diab: You can use the DAG, so you can see where the data is coming from and where it’s landing.
155 00:16:30.730 ⇒ 00:16:42.610 Hussein Diab: You can leverage dbt Cloud, you can leverage connections to many things. dbt Core is free, so you can integrate it anywhere you want.
156 00:16:43.620 ⇒ 00:16:47.019 Awaish Kumar: Okay, and what does the DBT seeds?
157 00:16:48.180 ⇒ 00:16:59.180 Hussein Diab: Yeah, you can use dbt seeds too, having flat files that will live somewhere, and you can use dbt build to do dbt seed, then to run and test at the same time.
158 00:16:59.550 ⇒ 00:17:13.540 Hussein Diab: You can do utils, there’s a lot of utilities that you can use. A lot of it will have to do with data quality. So, one of the clients that I’ve used,
159 00:17:13.540 ⇒ 00:17:29.990 Hussein Diab: that I worked with is DealNews, and they have a lot of data coming from different sources, and sometimes the data will be bad coming in from the source, so dbt test, for example, saved us a lot of time by catching these bad
160 00:17:29.990 ⇒ 00:17:35.149 Hussein Diab: Data coming in, and then we can do transformation to clean it up, make it better.
161 00:17:35.580 ⇒ 00:17:39.069 Awaish Kumar: So, what dbt tests you have written so far?
162 00:17:39.750 ⇒ 00:17:50.019 Hussein Diab: Uniqueness, not null, these are the most that I use. Uniqueness and not know.
163 00:17:51.610 ⇒ 00:17:58.100 Hussein Diab: sometime, like, I needed to have a specific format, it needs to be a specific regex format.
164 00:18:01.540 ⇒ 00:18:02.480 Awaish Kumar: Okay, have you ever…
165 00:18:02.480 ⇒ 00:18:04.060 Hussein Diab: Social, sir. Yeah.
166 00:18:04.400 ⇒ 00:18:06.640 Awaish Kumar: Have you ever used dbt Expectations?
167 00:18:09.050 ⇒ 00:18:11.150 Hussein Diab: I haven’t used dbt expectation.
168 00:18:12.470 ⇒ 00:18:18.350 Awaish Kumar: Okay, okay, I think,
169 00:18:19.690 ⇒ 00:18:26.220 Awaish Kumar: I’m good with all the technical things. We can… I just have a few last
170 00:18:26.360 ⇒ 00:18:46.050 Awaish Kumar: like, just, I think, one more question regarding how you communicate with the stakeholders. For example, especially non-technical stakeholders. So, for example, as a data person, we worked on something, we came up with some findings or some fact.
171 00:18:46.350 ⇒ 00:18:47.490 Awaish Kumar: Some facts.
172 00:18:47.670 ⇒ 00:19:02.469 Awaish Kumar: And… but when we send it to the non-technical stakeholder, which has the business domain knowledge, does not agree with our findings, how would you then pick your findings, or whatever you have, we have recommended?
173 00:19:04.410 ⇒ 00:19:09.899 Hussein Diab: So it’s all about clarity. It’s about speaking simple terms, non-technical terms.
174 00:19:11.780 ⇒ 00:19:17.050 Hussein Diab: We’ll provide them with examples, and I will communicate the risk.
175 00:19:17.240 ⇒ 00:19:21.170 Hussein Diab: If we go this way, this might happen.
176 00:19:21.300 ⇒ 00:19:25.979 Hussein Diab: I, like to communicate timeline.
177 00:19:26.750 ⇒ 00:19:29.690 Hussein Diab: I like to be proactive, so…
178 00:19:31.180 ⇒ 00:19:48.639 Awaish Kumar: I’m not talking about timelines, we are on time. We give them the findings. Like, he’s not a client, for example, he’s our internal stakeholder. We give him our findings, and he does not agree with the data. Like, if we are saying, okay, maybe our…
179 00:19:49.150 ⇒ 00:19:53.649 Awaish Kumar: Oh, Revenue for this specific product was of…
180 00:19:53.860 ⇒ 00:19:58.469 Awaish Kumar: Was, off by a few percentage points, and…
181 00:19:58.580 ⇒ 00:20:14.050 Awaish Kumar: And if he does not agree, or something like that, which is complex, like, revenue is still, like, something that you can easily show in number, in the data, in the invoices. But if you think, like, there is some,
182 00:20:14.140 ⇒ 00:20:31.610 Awaish Kumar: something, which is hard to convey and, like, hard to see through the numbers, then how we are going to basically communicate with our non-technical stakeholders that… that… that things that we have worked on are actually true.
183 00:20:31.610 ⇒ 00:20:39.760 Awaish Kumar: Because in the business, there are a lot of things that are sometimes assumed, because they thought it has been happening.
184 00:20:39.850 ⇒ 00:20:46.589 Awaish Kumar: In the market, and it might be happening the same way, but when you actually look at the data, it shows something different.
185 00:20:47.350 ⇒ 00:21:05.709 Awaish Kumar: Completely opposite of what they were thinking. So, they won’t easily trust the data and trust our findings. Then how… what will be your ways to convince them that actually our findings are much more accurate than… than what they were thinking?
186 00:21:06.760 ⇒ 00:21:19.770 Hussein Diab: I would walk them through the journey of what I’ve done, and I would use simpler terms. I would use non-technical words, I would use examples, like real-time examples, something about cars.
187 00:21:20.270 ⇒ 00:21:23.980 Hussein Diab: You know, or something that everyone is aware on how to do.
188 00:21:24.180 ⇒ 00:21:30.810 Hussein Diab: And, I will walk them through how I started from my sources.
189 00:21:31.070 ⇒ 00:21:42.809 Hussein Diab: what kind of transformation I did and why I did it, and I will walk them through the end result. I will state all of the assumptions, I will state all of the known facts, what’s known to be true.
190 00:21:43.560 ⇒ 00:21:46.600 Hussein Diab: okay.
191 00:21:48.070 ⇒ 00:21:56.930 Awaish Kumar: Okay, I think, yeah, that’s… that’s all from my side. I will leave last few minutes for you to ask any questions.
192 00:21:57.360 ⇒ 00:21:59.890 Hussein Diab: Yeah, so…
193 00:22:01.050 ⇒ 00:22:15.629 Hussein Diab: Kayla reached out to me on LinkedIn, and she told me, hey, I have this position, and then she set me up with you. So I know nothing about the position, I know nothing about… I tried to do just some…
194 00:22:15.800 ⇒ 00:22:27.009 Hussein Diab: you know, research online, but can you tell me more about the company, about what you guys do, what type of customers, what type of consulting, how big it is? Yeah.
195 00:22:27.510 ⇒ 00:22:34.089 Awaish Kumar: Okay, sure. So, Brainford is a data and AI consultancy firm.
196 00:22:34.260 ⇒ 00:22:44.690 Awaish Kumar: We provide, our services to mid to, like, large-scale enterprises, and
197 00:22:44.800 ⇒ 00:22:49.980 Awaish Kumar: Mostly, our… right now, our clients are in United States.
198 00:22:50.220 ⇒ 00:22:54.790 Awaish Kumar: And, yeah, we have different, like, data engineering.
199 00:22:54.990 ⇒ 00:23:12.480 Awaish Kumar: like, the domain, and we also have AI team, which provides AI services. And also, in data team, we also have different work streams, like data engineering, analytics engineering, data analysis. So we have the team for each of these different services.
200 00:23:12.710 ⇒ 00:23:16.799 Awaish Kumar: And I’m basically kind of lead data engineering.
201 00:23:16.920 ⇒ 00:23:24.520 Awaish Kumar: And we have my colleagues that lead different other work streams, data and analytics engineering and data analysis.
202 00:23:24.690 ⇒ 00:23:29.560 Awaish Kumar: Apart from that, in terms of the…
203 00:23:30.960 ⇒ 00:23:38.539 Awaish Kumar: The people, like, how we work, like, we are… we work remotely, everybody works, in their own
204 00:23:39.150 ⇒ 00:23:43.620 Awaish Kumar: Time zone, and we call up, like, we overlap for, like, maybe…
205 00:23:43.920 ⇒ 00:23:56.710 Awaish Kumar: few hours, so that we can actually communicate with each other, and we can communicate with clients. But apart from that, we are, like, free to work on our flexible hours, and…
206 00:23:57.760 ⇒ 00:24:03.690 Awaish Kumar: We have people from across the world, like, we have a lot of people from for example, Philippine.
207 00:24:03.830 ⇒ 00:24:10.029 Awaish Kumar: from… India, we have people from Europe, and also from US.
208 00:24:10.530 ⇒ 00:24:22.229 Awaish Kumar: So it’s… and yeah, the focus is on ascend communication, so what we want is actually the people should utilize Slack, or Zoom… Zoom clips, and,
209 00:24:22.420 ⇒ 00:24:31.449 Awaish Kumar: Notion, or GitHub, Write document, playbooks, and everything, so that, We can assign…
210 00:24:31.610 ⇒ 00:24:40.590 Awaish Kumar: Lee can asynchronously can communicate with each other, share updates, and stay on top of the things. We use Linear as our project management tool.
211 00:24:41.010 ⇒ 00:24:44.430 Awaish Kumar: And then Slack, obviously, for our communication.
212 00:24:44.700 ⇒ 00:24:51.090 Awaish Kumar: Oh… Yeah, that’s basically the kind of… the way we work at Brainforge.
213 00:24:51.230 ⇒ 00:25:10.480 Awaish Kumar: for the take-stake, like, it’s all, you know, like, it depends on the client-to-client basis, and their use cases. Mostly, we are the ones deciding on the tooling, but we don’t go with, like, okay, I know BigQuery, and so that’s why I’m just going to use BigQuery. We are more like, okay.
214 00:25:10.590 ⇒ 00:25:18.640 Awaish Kumar: client needs… these are the client needs, and then basically we map it to the tools that, okay, the BigQuery going to fit into this
215 00:25:18.840 ⇒ 00:25:21.030 Awaish Kumar: use case, or a snowflake, or…
216 00:25:21.230 ⇒ 00:25:27.140 Awaish Kumar: Or maybe use Azure for that, or whatever, like, Databricks, or… AF.
217 00:25:27.360 ⇒ 00:25:31.720 Awaish Kumar: Something like that. So it completely depends on, on that.
218 00:25:31.930 ⇒ 00:25:40.839 Awaish Kumar: Absolutely varies by… from client to client. So, dbt is kind of common tool, which is used across clients, so we… we…
219 00:25:41.000 ⇒ 00:25:47.960 Awaish Kumar: we can… we might be bringing in data from multiple sources to BigQuery to… To Snowflake.
220 00:25:48.220 ⇒ 00:26:05.150 Awaish Kumar: But when we employ that pattern that, okay, like, that the transformation that we do for our analytics, that’s all going to happen using dbt. I… so if your data is in BigQuery, Snowflake, but wherever that shift.
221 00:26:05.250 ⇒ 00:26:09.810 Awaish Kumar: We are going to use a dbt on top of it to… for all of our writing, our models.
222 00:26:09.990 ⇒ 00:26:13.699 Awaish Kumar: And then the PI trees tool is also…
223 00:26:13.960 ⇒ 00:26:24.300 Awaish Kumar: obviously client… there’s a lot of C from the client on BI tools, because those are the ones that the client interacts with, right?
224 00:26:24.440 ⇒ 00:26:26.379 Awaish Kumar: That is the tool that…
225 00:26:26.500 ⇒ 00:26:39.390 Awaish Kumar: many people in the client team are going to use. So, basically, they have a lot of say in that tooling, but the other parts of the infrastructure, we are basically going to recommend what they should be using.
226 00:26:39.710 ⇒ 00:26:52.039 Awaish Kumar: And for BI tool also, we have recommendations, but obviously, they need more, like, demos and things like that to actually see what, what looks good to them.
227 00:26:52.710 ⇒ 00:27:01.939 Hussein Diab: Yeah, this makes sense. Thank you. What can you tell me about the vision, the mission, and the growth of the company?
228 00:27:02.800 ⇒ 00:27:16.530 Awaish Kumar: Obviously the… it’s a… it’s a startup, so it’s a growth… in a growth stage. It’s a startup, we are growing, so we are, like… when I joined, basically, we were just maybe 5 to 8 people.
229 00:27:16.730 ⇒ 00:27:22.590 Awaish Kumar: Now it has been… it has grew to become maybe around 30 people right now.
230 00:27:22.740 ⇒ 00:27:31.470 Awaish Kumar: Which includes every… like, All the teams, like marketing, sales, dino.
231 00:27:31.550 ⇒ 00:27:48.359 Awaish Kumar: inside data, there are different work streams. Then we have an AI team, so all… so the full cloud… like, we have 30 people right now in the company, and it’s growing aggressively. We are hiring right now, maybe 4 or 5 more people in different work streams.
232 00:27:48.410 ⇒ 00:27:55.170 Awaish Kumar: And that will be higher in this quarter, maybe. So we are, like, growing a lot faster.
233 00:27:55.240 ⇒ 00:28:01.409 Awaish Kumar: The vision is that, like, to serve our clients,
234 00:28:03.040 ⇒ 00:28:07.809 Awaish Kumar: In a way that most of our… most… most of the consulting firms
235 00:28:07.980 ⇒ 00:28:16.820 Awaish Kumar: don’t work. Like, it’s a different way we work with them. We have different engagements, so we don’t assign, like, individual people just working on one client.
236 00:28:16.900 ⇒ 00:28:31.639 Awaish Kumar: We optimize our hours, like, if a client needs just 20 hours of a data engineer, we will just give that, and use our 20 hours from that engineer to maybe some other client. And
237 00:28:32.360 ⇒ 00:28:41.089 Awaish Kumar: Also, we use a lot of AI for our development, so we speed that up, so we basically… the people, maybe, the…
238 00:28:41.660 ⇒ 00:28:49.360 Awaish Kumar: existing big consultations. If they are doing the same thing in 3 months, we might be… we might be able to deliver it in a month.
239 00:28:49.410 ⇒ 00:28:53.519 Awaish Kumar: So we employ a lot of AI, we have our internal platform.
240 00:28:53.530 ⇒ 00:29:11.930 Awaish Kumar: Which helps us speed up things, like, if you communicate with the client on a meeting, so we have internal tools to basically, okay, use that, task script and the context for the client, and then head plus create linear tickets, and it’s just all there. You can just approve it, and it will create
241 00:29:12.090 ⇒ 00:29:23.500 Awaish Kumar: All the tickets in linear for you, along with all the assignments, like projects, and… yeah, and deadlines and things like that.
242 00:29:23.650 ⇒ 00:29:32.919 Awaish Kumar: And, basically, yeah, so it speeds up our development, it can help us create Gantt charts, so we have basically
243 00:29:33.400 ⇒ 00:29:40.319 Awaish Kumar: try to use AI to improve our workflows, and we are continue… we are continuing to do that.
244 00:29:40.520 ⇒ 00:29:44.989 Awaish Kumar: And that’s the unique thing about
245 00:29:45.110 ⇒ 00:30:01.359 Awaish Kumar: brain forge that it is, I’m just being… been here for one year, so, you know, in a year, we have grown, from 8 people to 30, so I am thinking… thinking, like, we are… we are going to grow it from here, really, aggressively, like.
246 00:30:01.720 ⇒ 00:30:10.000 Awaish Kumar: And, you know, maybe… In two years, like, it will be a… A lot better position.
247 00:30:10.320 ⇒ 00:30:11.240 Awaish Kumar: Oh, man.
248 00:30:12.180 ⇒ 00:30:13.070 Hussein Diab: Awesome.
249 00:30:13.420 ⇒ 00:30:15.899 Hussein Diab: Thank you. Thank you for answering me.
250 00:30:16.530 ⇒ 00:30:17.090 Awaish Kumar: Yeah, no worries.
251 00:30:17.090 ⇒ 00:30:19.379 Hussein Diab: We’re running out of time,
252 00:30:19.690 ⇒ 00:30:23.490 Hussein Diab: What’s the next step? When should I expect hearing back from you guys?
253 00:30:24.060 ⇒ 00:30:31.969 Awaish Kumar: for the next steps, like, after I submit my feedback, Our, like, the recruiter.
254 00:30:32.370 ⇒ 00:30:35.220 Awaish Kumar: Kayla, she’s going to basically come back
255 00:30:35.540 ⇒ 00:30:42.650 Awaish Kumar: To you with the next steps. Most… but our next steps are mostly, like, a meeting with another one of my colleagues.
256 00:30:42.820 ⇒ 00:30:49.689 Awaish Kumar: And then, we have maybe a home… take home test, and then final panel interview.
257 00:30:50.060 ⇒ 00:30:55.000 Awaish Kumar: Yeah, after that, if everything goes well, there will be an offer.
258 00:30:55.460 ⇒ 00:30:56.389 Awaish Kumar: And that’s all.
259 00:30:58.020 ⇒ 00:30:59.160 Hussein Diab: Sounds good.
260 00:30:59.910 ⇒ 00:31:02.109 Awaish Kumar: Okay, thank you for your time.
261 00:31:03.450 ⇒ 00:31:07.439 Awaish Kumar: Yeah, we are ready to get back to you soon. Thank you.
262 00:31:07.830 ⇒ 00:31:09.170 Hussein Diab: Sounds good. Thank you.