Meeting Title: Brainforge Interview w- Awaish Date: 2026-03-11 Meeting participants: Alejandro Gonzalez Sanchez, Awaish Kumar
WEBVTT
1 00:02:54.560 ⇒ 00:02:55.350 Awaish Kumar: Hi.
2 00:02:57.850 ⇒ 00:03:02.180 Alejandro Gonzalez Sanchez: Kodiphan is your name, sorry.
3 00:03:02.180 ⇒ 00:03:05.160 Awaish Kumar: Yeah, Avesh Kumar, what’s your name?
4 00:03:05.160 ⇒ 00:03:05.690 Alejandro Gonzalez Sanchez: Okay.
5 00:03:06.100 ⇒ 00:03:07.690 Alejandro Gonzalez Sanchez: Alejandro.
6 00:03:09.040 ⇒ 00:03:10.010 Awaish Kumar: Alejandro?
7 00:03:10.470 ⇒ 00:03:11.100 Alejandro Gonzalez Sanchez: Yeah.
8 00:03:11.480 ⇒ 00:03:13.899 Awaish Kumar: Okay. And where are you located?
9 00:03:14.780 ⇒ 00:03:20.380 Alejandro Gonzalez Sanchez: I’m located in Colombia, in the city of Medellin. Maybe… do you know this country?
10 00:03:21.550 ⇒ 00:03:26.479 Awaish Kumar: Colombia… Is it, yeah, it…
11 00:03:27.010 ⇒ 00:03:30.630 Awaish Kumar: Is it in a, like, correct.
12 00:03:30.630 ⇒ 00:03:33.060 Alejandro Gonzalez Sanchez: America. Yeah, exactly.
13 00:03:34.840 ⇒ 00:03:41.220 Awaish Kumar: Okay, yeah, I had, like, one of the colleagues before, At a different,
14 00:03:41.620 ⇒ 00:03:46.000 Awaish Kumar: place where, yeah, someone… we had someone from Colombia, So.
15 00:03:46.000 ⇒ 00:03:46.750 Alejandro Gonzalez Sanchez: Okay.
16 00:03:48.550 ⇒ 00:03:56.000 Awaish Kumar: Okay, so in this interview, it’s an introductory session where we are just going to talk about, .
17 00:03:57.480 ⇒ 00:04:03.530 Awaish Kumar: The… the, like, your… like, the… yourself, and, like, the… what you have been doing so far, what kind of…
18 00:04:03.710 ⇒ 00:04:11.170 Awaish Kumar: Projects you have worked on, and just get a deeper sense of, of your overall experience.
19 00:04:12.030 ⇒ 00:04:12.660 Alejandro Gonzalez Sanchez: Okay.
20 00:04:12.660 ⇒ 00:04:18.130 Awaish Kumar: So, yeah, my name is Avish Kumar, and I’m leading data engineering at Brainforce.
21 00:04:18.690 ⇒ 00:04:26.870 Awaish Kumar: And, you know, like, Brainforge is a consultancy which provides data and AI consultancy services to
22 00:04:26.980 ⇒ 00:04:38.159 Awaish Kumar: to the, mid to large enterprises in the… most… most of our clients right now are in U.S, and we operate remotely, so everybody works,
23 00:04:38.460 ⇒ 00:04:40.359 Awaish Kumar: From their own time zones.
24 00:04:40.470 ⇒ 00:04:43.359 Awaish Kumar: And we have employees from all over the world.
25 00:04:43.470 ⇒ 00:04:48.200 Awaish Kumar: So, yeah, getting started with… let’s get started with your introduction.
26 00:04:49.550 ⇒ 00:04:57.159 Alejandro Gonzalez Sanchez: Okay, as you know, my name is Alejandro Gonzalez-Sanchez. I studied the bachelor’s degree of computer science.
27 00:04:58.320 ⇒ 00:05:22.700 Alejandro Gonzalez Sanchez: Talking about my work experience, I started working in a… in a company from Spain. I used to create, like, backends to process some time series data there. Then I moved into a new company from Mexico. In that company, I just started to use, like, AWS services. I created some good jobs, Lambda… Lambdas in general.
28 00:05:22.700 ⇒ 00:05:31.320 Alejandro Gonzalez Sanchez: I handled with the data lake that they had in the company, because we have, like, several layers in which we will process the data.
29 00:05:31.640 ⇒ 00:05:44.850 Alejandro Gonzalez Sanchez: I also had to manage there some, you know, SQL databases, for example, Neptune and Redis, just because we had, like, an important project there. It was…
30 00:05:44.850 ⇒ 00:05:55.749 Alejandro Gonzalez Sanchez: Retrieving all the invoices from our clients, because we had to do, like, some kind of web scrapping process that will extract all the information.
31 00:05:55.750 ⇒ 00:06:02.739 Alejandro Gonzalez Sanchez: from the Mexican website for getting the invoices for the com… for the clients.
32 00:06:02.780 ⇒ 00:06:18.380 Alejandro Gonzalez Sanchez: we will do the web scrapping, and then we will store that in S3, and finally, with a snow pipe, we will ingest all that data inside of a table, or several tables in a snowflake.
33 00:06:18.880 ⇒ 00:06:24.160 Alejandro Gonzalez Sanchez: After that, my latest job is working at Tulio. In this
34 00:06:24.160 ⇒ 00:06:29.269 Alejandro Gonzalez Sanchez: I had the opportunity to also do some data modeling, but
35 00:06:29.270 ⇒ 00:06:51.669 Alejandro Gonzalez Sanchez: In addition to that, I have worked with the CI-CD pipelines. So, for example, I have some knowledge using orchestration tools like BuildKite to handle the CI-CD pipeline. Terraform, I started to use dbt just to create all the models that we require to… for our data lake.
36 00:06:52.020 ⇒ 00:07:11.540 Alejandro Gonzalez Sanchez: One important project there is one related with the migration of several databases that they were in MySQL and Postgres and so on. We wanted to migrate them, all of them, into our data lake that would have, like, a mailing architecture.
37 00:07:11.860 ⇒ 00:07:29.990 Alejandro Gonzalez Sanchez: So, that was an ongoing process that will require us to use, like, dbt, the service from AWS called EMR, just to have, like, SPARC processing behind dbt, and so on. So, those are, like, the main things that I could mention right now.
38 00:07:31.500 ⇒ 00:07:37.219 Awaish Kumar: Okay, so let’s dive into one of your recent projects.
39 00:07:37.220 ⇒ 00:07:38.379 Alejandro Gonzalez Sanchez: Yes. If you can…
40 00:07:38.380 ⇒ 00:07:43.179 Awaish Kumar: Explain me a little bit more about that. What exactly you did?
41 00:07:43.420 ⇒ 00:07:45.829 Awaish Kumar: What was the team structure? How…
42 00:07:46.560 ⇒ 00:07:54.849 Awaish Kumar: Like, the pipeline looked like, how you… how did you contribute it in that project, and what was the end result of that project?
43 00:07:56.240 ⇒ 00:08:12.050 Alejandro Gonzalez Sanchez: Okay, from my latest project, the one that I was telling you about, the migration from some databases to our data lake, I appeared into the picture in the middle of the project, because this is, like, well, it’s like an ongoing project.
44 00:08:12.150 ⇒ 00:08:22.240 Alejandro Gonzalez Sanchez: In which, the part that I was helping, first of all, I had to create several, models for the… for the tallies, depending on the…
45 00:08:22.420 ⇒ 00:08:25.720 Alejandro Gonzalez Sanchez: Requirements and some new tables that were appearing.
46 00:08:25.830 ⇒ 00:08:40.059 Alejandro Gonzalez Sanchez: I also contributed to this project to handle, for example, we need… I remember really well that we needed to have, like, the data from production to be testable from the development environment.
47 00:08:40.059 ⇒ 00:08:49.160 Alejandro Gonzalez Sanchez: So, one thing that I contributed there, it was creating, like, a small workflow just to move that data from the production environment
48 00:08:49.160 ⇒ 00:09:12.249 Alejandro Gonzalez Sanchez: to some kind of instance that the development environment could interact with that. The only problem with that is that you don’t want to move all the production data to the development environment, because you are sharing, like, some secrets that shouldn’t be shared. So what we did, it was, like, a simple process just to hash the information that could have, like.
49 00:09:12.250 ⇒ 00:09:32.970 Alejandro Gonzalez Sanchez: PII data. So, together with the data governance team, we will hash all the PII columns that we would consider that they shouldn’t be shared with the development account. For example, I don’t know, like, the cell phone of a user, or the… the sales that they have, some things like that.
50 00:09:32.970 ⇒ 00:09:37.390 Alejandro Gonzalez Sanchez: So, that was, like, a small part, but then I contributed with
51 00:09:37.390 ⇒ 00:09:39.670 Alejandro Gonzalez Sanchez: Several more pieces with that project.
52 00:09:40.470 ⇒ 00:09:51.710 Alejandro Gonzalez Sanchez: Now, the general infrastructure, because you… you asked for this, like, how that works, it would be, like, we will use… well, we are using, like, Meltano.
53 00:09:51.710 ⇒ 00:10:08.610 Alejandro Gonzalez Sanchez: We… to extract the information from the… from the sources, like, from the databases. So with Miltanum, we will just, like, have, like, a target that will… well, attack on a target, that they will, execute, like, a command to extract
54 00:10:08.610 ⇒ 00:10:12.979 Alejandro Gonzalez Sanchez: Incremental data from the… from the sources databases.
55 00:10:12.980 ⇒ 00:10:30.740 Alejandro Gonzalez Sanchez: Then, with that command, we will store everything in, like, in some kind of bronze layer in our data lake that obviously doesn’t have any kind of transformation. This will be executed by an… well, we have, like, an instead function that will execute, like, an ECS Fargate
56 00:10:30.740 ⇒ 00:10:31.680 Alejandro Gonzalez Sanchez: in which…
57 00:10:31.680 ⇒ 00:10:55.239 Alejandro Gonzalez Sanchez: inside of that, it was stored, like, Meltano. So with the step function, we will execute that part of Meltano. That will extract everything. After that, we will use also ECS just to execute our dbt models, just to start processing all the information, and depending on the table, we will execute, like, a command or another inside of dbt.
58 00:10:55.570 ⇒ 00:10:58.520 Alejandro Gonzalez Sanchez: That, then, what is oily, sorry.
59 00:10:58.520 ⇒ 00:11:03.349 Awaish Kumar: Let’s discuss that, how… so Meltano, we use that.
60 00:11:04.000 ⇒ 00:11:10.859 Awaish Kumar: To bring the data from sources to the data lake, that is, what your data lake is.
61 00:11:12.240 ⇒ 00:11:31.220 Alejandro Gonzalez Sanchez: What was the data like? It is, like, compa… like, many, packet files that were, that we were using, like, the, Mehalion architecture to store them, so we will have, like, the bronze layer in which we will store all the data, and also all the other, like.
62 00:11:31.220 ⇒ 00:11:33.990 Awaish Kumar: Question is, what do you use for your data lake?
63 00:11:35.480 ⇒ 00:11:43.939 Alejandro Gonzalez Sanchez: What do you mean? Like, how to handle with that, with, like, formation, for example, or what do you mean? Can you explain that, please?
64 00:11:43.940 ⇒ 00:11:52.770 Awaish Kumar: I mean, if I tell you, like, what database you used for restoring your structured data, you will come up with some names, like Postgres.
65 00:11:53.150 ⇒ 00:11:58.999 Awaish Kumar: SQL Server, or… MySQL… so here, my question is.
66 00:11:59.300 ⇒ 00:12:02.279 Awaish Kumar: For… for your data lake, what was used?
67 00:12:03.570 ⇒ 00:12:17.249 Alejandro Gonzalez Sanchez: Okay, so in this case, everything was being stored… well, we would have, like, the layer to query that data with Athena, but finally, this will be in… will be stored inside of Snowflake.
68 00:12:19.090 ⇒ 00:12:21.800 Awaish Kumar: When it comes from Beltano, then where it goes.
69 00:12:22.970 ⇒ 00:12:28.860 Alejandro Gonzalez Sanchez: It goes directly to S3, so we will just consult, like, query that data.
70 00:12:29.310 ⇒ 00:12:32.500 Awaish Kumar: Okay, and how’s… how was that,
71 00:12:32.720 ⇒ 00:12:35.750 Awaish Kumar: Once it is in S3, then you use,
72 00:12:36.210 ⇒ 00:12:39.699 Awaish Kumar: Like, what do you use to Snowflake?
73 00:12:41.470 ⇒ 00:12:52.800 Alejandro Gonzalez Sanchez: to connect it to Snowflake, we will do, like, all the transformation with dbt, so we will just query, well, execute all that data. We will use, like, the ingenu,
74 00:12:52.800 ⇒ 00:13:02.560 Alejandro Gonzalez Sanchez: the Office part, just to be able to execute that. That will do all the transformation layer by layer until we get to a snowflake to store that data there.
75 00:13:05.580 ⇒ 00:13:06.570 Awaish Kumar: Okay, so…
76 00:13:06.630 ⇒ 00:13:09.069 Alejandro Gonzalez Sanchez: Did you use… did you mention Spark?
77 00:13:10.220 ⇒ 00:13:10.810 Alejandro Gonzalez Sanchez: Yeah.
78 00:13:11.450 ⇒ 00:13:18.510 Awaish Kumar: So, like, how that flow was orchestrated, like, data is in S3,
79 00:13:18.650 ⇒ 00:13:24.060 Awaish Kumar: You are using a Spark to run some transformations.
80 00:13:24.160 ⇒ 00:13:27.720 Awaish Kumar: And then finally, that goes into the snowflake. Can you just…
81 00:13:27.900 ⇒ 00:13:31.160 Awaish Kumar: Let me explain, like, how that was happening.
82 00:13:32.010 ⇒ 00:13:55.700 Alejandro Gonzalez Sanchez: Yes, of course. So, we were using, like, DBT has several NGIs, for example, Athena. We were using the Spark one, in which we will require, like, an instance of Spark running. For that case, it could be a little bit difficult, because it was a little bit difficult to implement this in the company, but we were using, like, the Spark connector with an instance of Spark inside of
83 00:13:55.700 ⇒ 00:13:58.330 Alejandro Gonzalez Sanchez: the service EMR.
84 00:13:58.930 ⇒ 00:14:16.349 Alejandro Gonzalez Sanchez: So, with this service, we will just execute, like, from ECS, like, all these things, so that will, like, allow our DVT to use Spark as a, as an injoin. But before that, we were using just, like, the normal injoin of Athena.
85 00:14:17.790 ⇒ 00:14:21.350 Awaish Kumar: Okay, and then, how… where was the DVT involved?
86 00:14:22.490 ⇒ 00:14:30.870 Alejandro Gonzalez Sanchez: dbt was, like, the tool to process all the incoming data. So, for example, on the desktop.
87 00:14:30.870 ⇒ 00:14:34.940 Awaish Kumar: Yeah, yeah, but you were using Spark and dbt both?
88 00:14:35.760 ⇒ 00:14:36.360 Alejandro Gonzalez Sanchez: Yeah.
89 00:14:37.200 ⇒ 00:14:41.620 Alejandro Gonzalez Sanchez: Yeah, yeah. We were only using, like, the engine office pack with DVT.
90 00:14:44.230 ⇒ 00:14:48.680 Awaish Kumar: Like, I want to understand, like, Spark we use normally to do the transformation itself.
91 00:14:48.790 ⇒ 00:14:51.939 Awaish Kumar: So, why you are choosing Spark with dbt?
92 00:14:52.700 ⇒ 00:14:53.350 Awaish Kumar: Do you believe…
93 00:14:53.350 ⇒ 00:14:53.820 Alejandro Gonzalez Sanchez: Who knows?
94 00:14:54.040 ⇒ 00:14:59.330 Awaish Kumar: And Spark is also for transformation, so you are using the…
95 00:15:00.590 ⇒ 00:15:09.449 Awaish Kumar: both things for… at the same time, or different, like, where it was… how it was being used, I’m really…
96 00:15:09.560 ⇒ 00:15:11.710 Awaish Kumar: Confused, on that.
97 00:15:12.230 ⇒ 00:15:24.829 Alejandro Gonzalez Sanchez: Okay, let me explain. So, for example, in that case, so dbt is, like, obviously the transformation tool, but then you can have, like, several services, just like the engine behind of dbt.
98 00:15:24.830 ⇒ 00:15:36.820 Alejandro Gonzalez Sanchez: So, in this case, we will just have, like, an instance of EMR, of Spark that was ruining the dbt itself. So, it was taking advantage of the processing tools.
99 00:15:36.940 ⇒ 00:15:45.949 Alejandro Gonzalez Sanchez: Yeah. The only thing is that when they started to use, like, the, this Spark, this Spark in JAIN, or EMR,
100 00:15:45.950 ⇒ 00:15:56.959 Alejandro Gonzalez Sanchez: the… it happened before I appeared in the company, so I had to give, like, some kind of feedback or maintain the solution to try to keep improving those things, so yeah.
101 00:15:58.300 ⇒ 00:16:03.059 Awaish Kumar: But Spark Engine, basically, you can use to either write Python.
102 00:16:03.450 ⇒ 00:16:08.810 Awaish Kumar: Or you can write SQL, To transform your data.
103 00:16:09.470 ⇒ 00:16:11.300 Alejandro Gonzalez Sanchez: with Paisput, for example.
104 00:16:11.570 ⇒ 00:16:23.270 Awaish Kumar: I can write my Python script to read from S3 and do some transformation, and write it back, or something like that. Or I can use Spark SQL to do the same.
105 00:16:24.540 ⇒ 00:16:30.770 Awaish Kumar: I… yeah, I haven’t… Okay, yeah, we can move on.
106 00:16:31.070 ⇒ 00:16:39.830 Awaish Kumar: So… And so how do you rate yourself in, Dvd.
107 00:16:41.690 ⇒ 00:16:51.829 Alejandro Gonzalez Sanchez: Activity, I’m not… I don’t want to come in, like, myself, like, an expert. I have been using it… I’ve been… I’ve been using it for around… more or less than a year, so…
108 00:16:51.830 ⇒ 00:17:04.799 Alejandro Gonzalez Sanchez: if I… if you give me, like, an scale from 1 to 10, I would say, like, in general, maybe a 6, or in a good day, like, that 6. So, yeah, I’m still improving, but yeah, I have some knowledge in everything.
109 00:17:05.579 ⇒ 00:17:08.569 Awaish Kumar: Okay, what are seeds in DVD?
110 00:17:09.500 ⇒ 00:17:20.839 Alejandro Gonzalez Sanchez: The seeds are, like, files, like, that you can use, like, in good for other processes. They can be stored locally in your project, so you can access to them.
111 00:17:24.339 ⇒ 00:17:27.819 Awaish Kumar: Okay. So, it can be stored locally, and then what?
112 00:17:29.020 ⇒ 00:17:45.270 Alejandro Gonzalez Sanchez: And can be accessed from your models, or from anywhere else. They can be, for example, a CSV file, or Parquet, or whatever. So, they are, like, not necessarily, like, a table materialized, but just, existing in your project.
113 00:17:47.230 ⇒ 00:17:54.190 Awaish Kumar: Okay, and then, yeah, what are… Like.
114 00:17:55.200 ⇒ 00:18:00.559 Awaish Kumar: So I have to, use some custom naming.
115 00:18:01.080 ⇒ 00:18:06.200 Awaish Kumar: for my database and schemas. How can I do that, using dbt?
116 00:18:07.880 ⇒ 00:18:13.270 Alejandro Gonzalez Sanchez: For the custom naming, you can use, like, the reference macro inside of DVT.
117 00:18:13.460 ⇒ 00:18:26.219 Alejandro Gonzalez Sanchez: you can… it might… maybe when it doesn’t fail me, you can set that in the dbt project, so you can, like, reference those values, so they have, like, another value that you can use in your models in dbt.
118 00:18:28.930 ⇒ 00:18:33.500 Awaish Kumar: You mentioned about macros, like, what can I do in macros?
119 00:18:34.240 ⇒ 00:18:49.209 Alejandro Gonzalez Sanchez: You can solve some kind of… for example, if you want, you can add, like, some logic business to those micros. So, for example, if you only want to, I don’t know, like.
120 00:18:49.220 ⇒ 00:19:05.609 Alejandro Gonzalez Sanchez: You want to create, like, a function that, retrieves some kind of, information. For example, what columns you want to select, or that kind of thing. You can create that with a macro that you can then use for, yeah.
121 00:19:07.310 ⇒ 00:19:14.980 Awaish Kumar: Yeah, I can use a… I can write a macro and write some logic, but how that… how dbt will use my…
122 00:19:16.360 ⇒ 00:19:21.100 Awaish Kumar: Oh, for… Figuring out the names of the databases.
123 00:19:21.290 ⇒ 00:19:23.210 Awaish Kumar: And these datasets.
124 00:19:24.240 ⇒ 00:19:34.710 Alejandro Gonzalez Sanchez: So, it uses, like, Jinja, so you can use, like, the reference function just to do that. Is that what you are asking, sorry? Or what do you mean?
125 00:19:35.850 ⇒ 00:19:41.710 Awaish Kumar: So, like, whenever a dbt runs, it figures out the table’s name, schema.
126 00:19:42.030 ⇒ 00:19:45.739 Awaish Kumar: Based on the logic either you provided.
127 00:19:46.350 ⇒ 00:19:52.129 Awaish Kumar: Either it is, like, a standard, you provide in your profile.yml file.
128 00:19:52.370 ⇒ 00:19:57.890 Awaish Kumar: to what database, what schemas you should connect to, and write models. Or you provide it…
129 00:19:58.790 ⇒ 00:20:01.619 Awaish Kumar: All logic in a… as a macro.
130 00:20:01.770 ⇒ 00:20:04.390 Awaish Kumar: I’m trying to understand if I’m…
131 00:20:04.590 ⇒ 00:20:12.740 Awaish Kumar: if I’m writing a macro, what makes dbt to use my custom macro instead of reading it from
132 00:20:12.930 ⇒ 00:20:16.269 Awaish Kumar: like, profile.yml file, or anywhere else.
133 00:20:18.420 ⇒ 00:20:19.410 Alejandro Gonzalez Sanchez: So…
134 00:20:21.520 ⇒ 00:20:38.419 Alejandro Gonzalez Sanchez: So, yes, because obviously all the macros are stored… you can store locally your macros, or, like, use from any dependency, like dbt UTs, but in the case that you want to directly say to dbt, like, hey, use my database and my schema.
135 00:20:38.700 ⇒ 00:20:39.590 Alejandro Gonzalez Sanchez: Ehh…
136 00:20:42.820 ⇒ 00:20:43.740 Alejandro Gonzalez Sanchez: When…
137 00:20:43.830 ⇒ 00:20:58.960 Alejandro Gonzalez Sanchez: in that case, I would say, like, the activity profiles that obviously create, like, the connection with your instance, for example, Snowflake, or Podres, or whatever that you are using, you can configure there those kind of connections.
138 00:20:58.960 ⇒ 00:21:06.580 Alejandro Gonzalez Sanchez: in which you are able to connect to those places. I don’t know if that answered your question properly.
139 00:21:07.590 ⇒ 00:21:14.460 Awaish Kumar: Okay, that’s fine. I have a few more questions, like… Yes.
140 00:21:14.770 ⇒ 00:21:24.029 Awaish Kumar: In terms of, for example, if you are talking to a stakeholder.
141 00:21:24.330 ⇒ 00:21:27.159 Awaish Kumar: Or you could say you’re kind of talking to a client.
142 00:21:27.290 ⇒ 00:21:35.190 Awaish Kumar: How would you explain your… Findings, or… or… To a non-technical stakeholder.
143 00:21:36.380 ⇒ 00:21:52.850 Alejandro Gonzalez Sanchez: So, the best thing that you can do there is try to… like, data by itself is difficult to understand if you don’t have… if you don’t have, like, the right context. So, probably creating some kind of dashboard that gives some insights of what your data is…
144 00:21:53.020 ⇒ 00:22:05.989 Alejandro Gonzalez Sanchez: what is going on with your data? So, I will try to create that, try to be… use, like, a normal vocabulary without saying anything really complicated, just to
145 00:22:06.290 ⇒ 00:22:11.790 Alejandro Gonzalez Sanchez: the other person understands better what I’m trying to express.
146 00:22:11.980 ⇒ 00:22:21.329 Alejandro Gonzalez Sanchez: So, that’s one thing. It’s also important to understand the context of your stakeholders, so if you can communicate with boards that they understand, that’s even better.
147 00:22:21.770 ⇒ 00:22:25.869 Alejandro Gonzalez Sanchez: And yeah, I think those are the main things that I would have in mind.
148 00:22:26.850 ⇒ 00:22:27.640 Awaish Kumar: Okay.
149 00:22:27.910 ⇒ 00:22:30.340 Awaish Kumar: And, yeah, if you have,
150 00:22:30.570 ⇒ 00:22:36.480 Awaish Kumar: conflict in your team regarding any technical solution? How would you… I’ll answer that.
151 00:22:37.420 ⇒ 00:22:53.990 Alejandro Gonzalez Sanchez: This has happened to me in the past. I remember once, the team I… the… my team and I was… were creating, like, a project for the data science team. We wanted to refactor some things that they did.
152 00:22:53.990 ⇒ 00:22:58.659 Alejandro Gonzalez Sanchez: We did it, but in the end of the project, when we did, like, the demo.
153 00:22:58.660 ⇒ 00:23:14.850 Alejandro Gonzalez Sanchez: what we delivered, it wasn’t exactly what they wanted, so that caused many problems with the two teams. The solution there was obviously trying to communicate. Communication is, like, a key factor in any team.
154 00:23:15.170 ⇒ 00:23:19.019 Alejandro Gonzalez Sanchez: So, I… we will talk with the other team, like, hey.
155 00:23:19.020 ⇒ 00:23:38.770 Alejandro Gonzalez Sanchez: This is what we have, what you never were, like, in the meetings, but instead of doing this, now, in the future, we are gonna have you more in mind for the process that we are doing, so every spring, they will take part in the meetings, and we will show them, like, some… what the progress was until that point.
156 00:23:38.930 ⇒ 00:23:53.939 Alejandro Gonzalez Sanchez: So, that was, like, a solution for that problem, trying to communicate, be respectful, understanding that the other person also has some priorities in their list. By doing that, we had, like, a better understanding in that project.
157 00:23:55.070 ⇒ 00:23:55.790 Awaish Kumar: Okay.
158 00:23:56.000 ⇒ 00:23:58.729 Awaish Kumar: Okay, I just have one last question,
159 00:23:58.990 ⇒ 00:24:02.599 Awaish Kumar: Yes. Why are you, looking for a new role?
160 00:24:04.520 ⇒ 00:24:20.910 Alejandro Gonzalez Sanchez: To be honest, like, the current company, I’m right now working as a contractor, I don’t want to lie to you, because that’s not necessary, but in the current company, where I’m working, they… a layout happened, and they decided to…
161 00:24:20.910 ⇒ 00:24:28.629 Alejandro Gonzalez Sanchez: let’s say, fire all the contractors, and I was one of the contractors in the data team, so only that affect me.
162 00:24:28.630 ⇒ 00:24:35.839 Alejandro Gonzalez Sanchez: my position. I’m still working for the company, but I’m finalizing, like, my process within, like, this month.
163 00:24:37.220 ⇒ 00:24:42.380 Awaish Kumar: Okay, and what excited you to be part of Brainforce?
164 00:24:43.540 ⇒ 00:24:54.069 Alejandro Gonzalez Sanchez: I was reading about you in your website. As I can see, you work with multiple clients. I think that’s a good spinning, at least for me, because
165 00:24:54.080 ⇒ 00:25:07.539 Alejandro Gonzalez Sanchez: I… most of the time, I have been working creating solutions for any specific company, but I have never worked, like, directly with creating solutions for multiple clients. So that’s, for me, at least, like, an interesting challenge.
166 00:25:07.710 ⇒ 00:25:23.120 Alejandro Gonzalez Sanchez: I also was reading that you… well, there are some good metrics about the job that you have performed. I really want to be part of that, because I want to keep growing, like, my work experience, start learning more things,
167 00:25:23.120 ⇒ 00:25:29.620 Alejandro Gonzalez Sanchez: like, the focus, or, like, the way that you do your things, it’s… I find it at least interesting.
168 00:25:31.360 ⇒ 00:25:32.040 Awaish Kumar: Okay.
169 00:25:32.590 ⇒ 00:25:39.340 Awaish Kumar: I think that’s it from me. If you have any other questions for me… I can answer that.
170 00:25:40.530 ⇒ 00:25:52.780 Alejandro Gonzalez Sanchez: Yes, I was wondering, for example, if in the case that I get hired in your company, what would be the most interesting project that is on the way?
171 00:25:54.980 ⇒ 00:26:02.950 Awaish Kumar: Okay, so… being at Brainforge is interesting, because you’re going to get… get to work on…
172 00:26:03.210 ⇒ 00:26:06.710 Awaish Kumar: on, multiple… projects.
173 00:26:07.070 ⇒ 00:26:10.360 Awaish Kumar: At the same time, so we, we don’t…
174 00:26:10.530 ⇒ 00:26:27.529 Awaish Kumar: just assign, like, the person full-time to any client. It’s normally, time going to be split between multiple clients. Two or three, it could be any, it could be two, you know, it could be three, based on the…
175 00:26:28.710 ⇒ 00:26:36.759 Awaish Kumar: work… the help needed for that client. So, for example, if you are hired as a Data engineer, then…
176 00:26:36.900 ⇒ 00:26:44.000 Awaish Kumar: It depends on… one of our clients needs, like, more hours from a data engineer than the other one, so we are going to.
177 00:26:44.000 ⇒ 00:26:44.550 Alejandro Gonzalez Sanchez: I mean…
178 00:26:44.550 ⇒ 00:26:45.440 Awaish Kumar: accordingly.
179 00:26:45.740 ⇒ 00:26:54.410 Awaish Kumar: And we are… our text, it looks like, yeah, we are using dbt, Snowflake, Fivetran, Alatomic.
180 00:26:54.740 ⇒ 00:26:58.089 Awaish Kumar: These are some of the tools. We use Omni as a BI tool.
181 00:26:58.300 ⇒ 00:27:07.569 Awaish Kumar: And, yeah, some clients use GCP, some use AWS, it all depends on, on the client. And…
182 00:27:07.570 ⇒ 00:27:07.940 Alejandro Gonzalez Sanchez: Yes.
183 00:27:08.060 ⇒ 00:27:14.009 Awaish Kumar: Not just on the client that they come up with these tools, It depends on them.
184 00:27:15.180 ⇒ 00:27:30.800 Awaish Kumar: on their use cases, basically, that’s what I’m trying to say. So, they come up with their problem, their pain points, and we gather these solutions for them, and based on whatever recommendations are, we
185 00:27:32.860 ⇒ 00:27:52.370 Awaish Kumar: we also recommend the tool, right? And that you should use this, and based on that, yeah, we are involved in those decisions, but yes, it depends. So our… the tech stack is… varies from client to client, and it depends on the use cases of the client.
186 00:27:53.410 ⇒ 00:27:54.020 Alejandro Gonzalez Sanchez: King.
187 00:27:54.370 ⇒ 00:28:00.150 Awaish Kumar: Yeah, so, like, we have ongoing projects, for example, like, as I mentioned, one of the projects
188 00:28:00.270 ⇒ 00:28:11.220 Awaish Kumar: on AWS, we are bringing in data from some data lake into the Snowflake, and then we do dbt on top of it for processing.
189 00:28:11.820 ⇒ 00:28:20.649 Awaish Kumar: And there’s a lot of, like, challenges at different levels. Number one, like, we need to set up all that infra for the client, but the second thing is that you…
190 00:28:20.800 ⇒ 00:28:23.069 Awaish Kumar: I’ll have to work with the data, the data…
191 00:28:23.310 ⇒ 00:28:28.180 Awaish Kumar: It requires identity switching, data requires… there are some challenges at adult level.
192 00:28:28.330 ⇒ 00:28:35.299 Awaish Kumar: Where you need to clean it up, combine it with multiple sources.
193 00:28:35.450 ⇒ 00:28:41.139 Awaish Kumar: To create a golden dataset, so that’s… That’s what it is.
194 00:28:42.310 ⇒ 00:28:47.450 Alejandro Gonzalez Sanchez: Okay, I was wondering, for example, you told me that,
195 00:28:47.780 ⇒ 00:29:04.049 Alejandro Gonzalez Sanchez: obviously, like, the project and the tech stack depends on the client that approached you, so it’s possible that a client uses AWS and another uses ECP, so what do you do with your team in terms of that? Like, I mean, for example.
196 00:29:04.280 ⇒ 00:29:17.070 Alejandro Gonzalez Sanchez: Do you have, like, people who work in multiple, cloud, services, or is it, like, if you need to learn, like, a technology to help a client, you give, like, the.
197 00:29:17.070 ⇒ 00:29:17.390 Awaish Kumar: No, no.
198 00:29:17.390 ⇒ 00:29:17.800 Alejandro Gonzalez Sanchez: for you.
199 00:29:17.800 ⇒ 00:29:23.420 Awaish Kumar: We don’t have any… cloud-specific experts. We are all
200 00:29:23.540 ⇒ 00:29:27.009 Awaish Kumar: Working with, all the cloud environments.
201 00:29:28.180 ⇒ 00:29:29.080 Alejandro Gonzalez Sanchez: Okay.
202 00:29:29.080 ⇒ 00:29:36.020 Awaish Kumar: One could be… one could be more, like, have much more expertise. For example,
203 00:29:36.490 ⇒ 00:29:47.890 Awaish Kumar: a person A has more experience in working in GCP, or A has some certifications in GCP, obviously we are going to get his feedback, his guidance on those
204 00:29:48.140 ⇒ 00:29:58.050 Awaish Kumar: projects, where we require GCP. Same for any other cloud provider, but we don’t divide our tasks based on
205 00:29:58.470 ⇒ 00:30:06.970 Awaish Kumar: the tech stack. If you’re a data engineer, we expect you to just dive in into whatever tech stack there is.
206 00:30:08.070 ⇒ 00:30:18.239 Alejandro Gonzalez Sanchez: Yeah, that makes sense, and even more with technology, like ECP, AWS, and Azure, because most of them, well, they are really similar with each other, so it’s just trying to understand
207 00:30:18.500 ⇒ 00:30:23.650 Alejandro Gonzalez Sanchez: What service translates into what other service, and just starting to work with that.
208 00:30:24.240 ⇒ 00:30:28.709 Alejandro Gonzalez Sanchez: Okay, so, you also told me that…
209 00:30:28.920 ⇒ 00:30:43.029 Alejandro Gonzalez Sanchez: depending on the client, you will require, like, to put more effort, for example, for another time you need, or depending on the client, it could require less effort. So, how is, like, a normal day in your company? I’m just curious about that.
210 00:30:44.070 ⇒ 00:30:46.250 Awaish Kumar: I think a normal day is that you…
211 00:30:46.570 ⇒ 00:30:49.819 Awaish Kumar: You already know what clients you are assigned on.
212 00:30:49.980 ⇒ 00:30:57.970 Awaish Kumar: And then we have… We have people who create tickets for you, and then we’ll just work on.
213 00:30:58.480 ⇒ 00:31:10.779 Awaish Kumar: those tickets, and try to close out as much as possible, and be active on Slack, try to give updates. We have, regular… we don’t have… we have some regular meetings, between
214 00:31:11.380 ⇒ 00:31:19.030 Awaish Kumar: Like, the stand-ups between the… the… kind of a whole company leadership, where we talk about
215 00:31:19.260 ⇒ 00:31:27.260 Awaish Kumar: The client… Needs, but otherwise, it, it is depend… it is completely… our… our…
216 00:31:27.380 ⇒ 00:31:32.120 Awaish Kumar: Teams run completely independently, so it depends on the team to…
217 00:31:32.360 ⇒ 00:31:35.639 Awaish Kumar: It is completely on the team lead to decide on what,
218 00:31:36.250 ⇒ 00:31:44.090 Awaish Kumar: How he wants to handle his client, or if he wants to have regular stand-ups, or once a week, or a sing, whatever.
219 00:31:45.200 ⇒ 00:31:50.179 Alejandro Gonzalez Sanchez: Okay, I understand. So… We have one last… sorry.
220 00:31:50.640 ⇒ 00:31:51.600 Awaish Kumar: Yeah, go on.
221 00:31:52.320 ⇒ 00:32:03.099 Alejandro Gonzalez Sanchez: Sorry, I only have one last question, because I don’t want to take more of your time, honestly. Why is NES after this? It was out of curiosity?
222 00:32:04.230 ⇒ 00:32:05.269 Awaish Kumar: Sorry, what is?
223 00:32:05.670 ⇒ 00:32:08.609 Alejandro Gonzalez Sanchez: What is next? Like, what is after this meeting?
224 00:32:08.610 ⇒ 00:32:14.799 Awaish Kumar: The next is that I’m going to get… submit my feedback, and after that,
225 00:32:15.270 ⇒ 00:32:23.980 Awaish Kumar: Our recruiters will get back to you. Normally, we have, like, another interview, then, maybe, home-based task.
226 00:32:24.190 ⇒ 00:32:28.359 Awaish Kumar: And then a final revision of that, and then… And that’s it.
227 00:32:29.570 ⇒ 00:32:33.599 Alejandro Gonzalez Sanchez: Gotcha, gotcha. Okay, thank you so much for answering all my questions.
228 00:32:34.620 ⇒ 00:32:35.670 Awaish Kumar: Yep, thank you.
229 00:32:35.970 ⇒ 00:32:36.670 Awaish Kumar: Right?
230 00:32:37.410 ⇒ 00:32:37.880 Alejandro Gonzalez Sanchez: Thank you.