Meeting Title: Brainforge AI Engineer Intro Call Date: 2026-03-04 Meeting participants: Mouhamad, Kaela Gallagher
WEBVTT
1 00:01:25.040 ⇒ 00:01:26.939 Kaela Gallagher: Hi, Mohammed, how’s it going?
2 00:01:30.350 ⇒ 00:01:32.230 Mouhamad: Hi, Kayla, how are you?
3 00:01:32.660 ⇒ 00:01:33.600 Kaela Gallagher: Good!
4 00:01:34.180 ⇒ 00:01:40.669 Kaela Gallagher: Thanks so much for taking the time to meet with me. I hope that you’re doing okay, and you and your family are safe.
5 00:01:41.260 ⇒ 00:01:48.889 Mouhamad: Yes, thank you so much. Yeah, we have moved, like, up north, more up north. We are safe now, yes.
6 00:01:49.000 ⇒ 00:01:51.130 Mouhamad: It’s quite a bit.
7 00:01:52.370 ⇒ 00:01:58.010 Kaela Gallagher: Okay, okay, got it. Oh my gosh, what a crazy time. Yeah, we’re…
8 00:01:58.010 ⇒ 00:02:00.499 Mouhamad: Crazy time to be in the Middle East now, yeah.
9 00:02:00.890 ⇒ 00:02:18.140 Kaela Gallagher: Yeah, yeah, I can imagine. Well, I’m… I’m grateful that you have still made… made time to, chat with me, and excited to kind of learn a little bit more about you, and, yeah, see if any of our roles might kind of…
10 00:02:18.150 ⇒ 00:02:26.080 Kaela Gallagher: match what you’re looking for. But wanted to just start off by asking, like, why are… why are you looking for a new position?
11 00:02:27.280 ⇒ 00:02:27.810 Mouhamad: Yeah.
12 00:02:28.010 ⇒ 00:02:44.040 Mouhamad: So I have been in the current company that I’m in for about, now, three years, 3 and a half years, and the company that I am in is basically a data consultancy. So, we do work with a lot of, like, AI, data, everything.
13 00:02:44.200 ⇒ 00:02:45.960 Mouhamad: But it’s sort of like…
14 00:02:46.870 ⇒ 00:02:51.629 Mouhamad: I sometimes be thrown at, like, 3, 4, 5, 6 projects at the same time.
15 00:02:51.910 ⇒ 00:02:57.669 Mouhamad: So I want… I want to have, like, a new challenge where I’m working on, like, a…
16 00:02:57.770 ⇒ 00:03:00.680 Mouhamad: Tfter or product or something, like…
17 00:03:01.040 ⇒ 00:03:08.759 Mouhamad: It’s not that it’s, like, more chill, I know that it still works, especially in a startup, and I have worked in a startup before.
18 00:03:09.080 ⇒ 00:03:22.090 Mouhamad: But it would be, like, I would love it more to work on something better than just being for something that working with for other firms, not other firms, for other clients, etc.
19 00:03:23.090 ⇒ 00:03:30.519 Kaela Gallagher: Okay, okay, so potentially leaving consulting is what you’re saying? Or working with less clients?
20 00:03:31.660 ⇒ 00:03:36.929 Mouhamad: Okay, well, that’s… let’s take. That would be, like, a little bit killer.
21 00:03:38.310 ⇒ 00:03:41.839 Kaela Gallagher: Okay, okay, got it.
22 00:03:41.990 ⇒ 00:03:48.720 Kaela Gallagher: I’m curious, like, what is the most… you mentioned you’re working in data, but also AI?
23 00:03:49.630 ⇒ 00:03:50.250 Mouhamad: Yep.
24 00:03:50.250 ⇒ 00:03:51.490 Kaela Gallagher: Just the data, okay.
25 00:03:52.280 ⇒ 00:04:06.309 Mouhamad: No, no, no, no, both of them. So, in my company, we don’t have, like, a role like AI engineer, data scientist, no. So, a data scientist is also a data… an AI engineer is also…
26 00:04:06.450 ⇒ 00:04:21.019 Mouhamad: And, like, the newer age of LLM, LMM, engineering, so everything. So the data scientist does everything, from just the normal data science work to the actual AI, so from end to end, everything. So I do everything.
27 00:04:21.560 ⇒ 00:04:38.519 Kaela Gallagher: Okay, okay, cool. I’m curious, like, in… in this role, what has been your, like, proudest achievement? Like, a time that you built a solution that you were, really proud of, maybe it delivered, like, a lot of impact.
28 00:04:40.640 ⇒ 00:04:41.150 Mouhamad: Yep.
29 00:04:43.620 ⇒ 00:04:54.319 Mouhamad: I would say one of the proudest is… so, one of the clients that we worked with was Yod Air, so it’s an airline company.
30 00:04:56.700 ⇒ 00:05:10.620 Mouhamad: It’s a big airline company that opened in Saudi Arabia, like, a year ago or something, and we’re doing, like, AI for them and stuff. And they basically wanted, like, a… sort of a chatbot, but also does some actions.
31 00:05:10.830 ⇒ 00:05:16.969 Mouhamad: And that chatbot is not just, like, a normal chatbot, like, on text, etc, no, so they had, like.
32 00:05:17.440 ⇒ 00:05:27.299 Mouhamad: Like, airline images, they have, like, planes images, they have also all sorts of data, like, from documents to sketches to…
33 00:05:27.850 ⇒ 00:05:47.690 Mouhamad: to image… to a lot of things, because they wanted to do actions, so also the chatbot can do some actions, like, also, like, they can type it and do this for me, like, they can also, get information, they can also book on the calendar some stuff, so it was, like, entire system, and I was leading development for the team.
34 00:05:47.840 ⇒ 00:05:53.180 Mouhamad: I am proud of this week because, one, the deadline was very tight.
35 00:05:53.430 ⇒ 00:05:55.710 Mouhamad: So we had to deliver quickly.
36 00:05:55.960 ⇒ 00:06:13.509 Mouhamad: Also because the data wasn’t, like, anything usual. Like, usually for the chatbot, it’s usually text, usually some images, like these things. No, for this one, it was all kinds of data, all kinds of messy data.
37 00:06:13.650 ⇒ 00:06:26.959 Mouhamad: So there was a lot of work, and I was leading development with the team, and etc. I’m proud of it, because it was very well done at the end. We have delivered something very good. But it was hectic at the same time, because of the timeline kind of declined.
38 00:06:27.160 ⇒ 00:06:33.360 Mouhamad: Also, like, they kept going back and forth, they want, amendments and stuff. So yeah, that’s one of the projects.
39 00:06:33.830 ⇒ 00:06:47.270 Kaela Gallagher: Awesome. So, what about a situation where, like, one of your models or your AI tools was underperforming? How do you kind of, like, diagnose and address an issue like that?
40 00:06:48.060 ⇒ 00:06:48.790 Mouhamad: Yep.
41 00:06:49.110 ⇒ 00:06:59.780 Mouhamad: So, if, let’s say, a model is underperforming or anything, the first thing that I would do is, I would create, like, a data set, so the data set in the evaluation set, for example.
42 00:06:59.890 ⇒ 00:07:06.969 Mouhamad: And I would do some running samples on this evaluation, and I would check what sort of metrics I’m getting.
43 00:07:07.130 ⇒ 00:07:17.709 Mouhamad: If it’s underperforming on some aspect, there are some methods. So, for example, I would first go to the right prompt. So, I would do some prompt engineering, fixing the problem.
44 00:07:17.850 ⇒ 00:07:24.310 Mouhamad: Checking some… maybe to do some chain of thoughts in the realms, etc, some techniques.
45 00:07:24.460 ⇒ 00:07:36.329 Mouhamad: That’s the first thing. Second thing, which is also trivial, but a lot of the people miss it, maybe move to another model. Maybe that model is not suitable for this kind of position, or this kind of problem, sorry.
46 00:07:36.450 ⇒ 00:07:46.480 Mouhamad: A lot of people will be like, oh, no, you can use the model. No, no, it depends on the model, on the problem, on everything. It’s not just, like, use GPT and it will work.
47 00:07:46.800 ⇒ 00:08:05.950 Mouhamad: Another thing could be fine-tuning, which is very important. So a lot of also the people, like, would use, like, an LLM, but it’s big, and it’s on many domains, they can understand everything, but no, some sort of problem needs fine-tuning on a specific domain, so it performs well on that domain.
48 00:08:06.280 ⇒ 00:08:12.629 Mouhamad: That is also some of the techniques, for example. It depends on the problem, on the metrics that we get.
49 00:08:13.280 ⇒ 00:08:22.450 Kaela Gallagher: Yeah, okay, okay, cool. Well, I know I had reached out, I think it was specifically in regard to, like, our AI engineering position.
50 00:08:22.450 ⇒ 00:08:22.870 Mouhamad: Yes.
51 00:08:22.870 ⇒ 00:08:41.530 Kaela Gallagher: So, just to give a little bit more information into Brainforge and what we do, we are a data, like, consultancy as well. So, we partner with clients, some of them are kind of, like, consumer good and e-com brands. We have, some, like.
52 00:08:41.630 ⇒ 00:09:00.350 Kaela Gallagher: health, clients, we’re starting to work with, like, financial clients as well, so, you know, you would be working with, probably a couple clients at once, supporting AI efforts, and then we also have a really kind of robust, like, internal
53 00:09:00.350 ⇒ 00:09:05.989 Kaela Gallagher: AI system that sometimes our AI engineers support as well.
54 00:09:06.310 ⇒ 00:09:19.890 Kaela Gallagher: So that’s an… that’s an overview of us, what the interview process would look like is it’s 3 rounds after this. So I’m on the recruiting side, this is just kind of like an intro call, I can help answer any questions.
55 00:09:20.050 ⇒ 00:09:36.820 Kaela Gallagher: Yeah, the next steps would be, just kind of like an initial, culture fit interview, with one of our team members, Sam. Then you would chat with Pranav, for more, like, deep technical interview.
56 00:09:36.820 ⇒ 00:09:49.449 Kaela Gallagher: And then our final round is, like, a challenge that we’d have you complete, and, you would present that to… to a panel, and they would kind of walk through your solution. So, that’s what next steps look like here.
57 00:09:49.450 ⇒ 00:10:00.030 Kaela Gallagher: we could complete all three of those rounds in the next couple weeks, give or take. So, yeah, that’s kind of our timeline. Curious if you have any questions.
58 00:10:01.590 ⇒ 00:10:10.330 Mouhamad: So I would like to understand first, I know, I know it says, like, on the form that it says startups, so how many people are in the startup so far?
59 00:10:10.510 ⇒ 00:10:22.180 Mouhamad: How many clients, potentially, that you have, what sort of also, like, until you said that there’s internal, also, like, solution that they’re doing or something, what is it about, exactly?
60 00:10:23.480 ⇒ 00:10:34.110 Kaela Gallagher: Yes. So, in terms of our team size, we’re about 25 team members right now, but actively growing. We have…
61 00:10:34.330 ⇒ 00:10:36.850 Kaela Gallagher: I would say, probably…
62 00:10:36.920 ⇒ 00:10:57.879 Kaela Gallagher: about 10 active clients right now. Our typical engagement is about 3 months long with our clients. We do have a couple that we’ve been working with for, you know, over a year, and we kind of function as, like, a data team for them. But a lot of our engagements are going to be more around the 3-month mark, so our number of clients is…
63 00:10:57.880 ⇒ 00:11:05.560 Kaela Gallagher: Always kind of changing. And then in terms of our internal system,
64 00:11:06.370 ⇒ 00:11:12.350 Kaela Gallagher: Right now, what I use it for the most on, like, the recruiting and people side of the house.
65 00:11:12.350 ⇒ 00:11:36.079 Kaela Gallagher: is, it basically, like, records our Zoom calls, and we have a repository of, like, basically our company brain, and I can go ask it for, you know, notes on our call today, or I could ask it to, you know, hey, I’m meeting this team member for the first time, can you tell me a little bit about them? And it’s gonna have some information on the team member.
66 00:11:36.080 ⇒ 00:11:45.889 Kaela Gallagher: And it’s kind of just, like, the brain of our company. But, I’m sure, like, one of our data team members could give you, like, an overview of it in a way more.
67 00:11:45.890 ⇒ 00:11:46.550 Mouhamad: It’s ridiculous.
68 00:11:46.550 ⇒ 00:11:48.449 Kaela Gallagher: then, too. Yeah.
69 00:11:48.450 ⇒ 00:11:49.080 Mouhamad: I hope.
70 00:11:49.400 ⇒ 00:11:52.719 Mouhamad: Yes, I think these are the questions that I had, yeah.
71 00:11:53.170 ⇒ 00:12:10.079 Kaela Gallagher: Okay, cool. Well, if you’re interested, I can go ahead and get you, set up for a first round. I’ll go ahead and send you an email with, like, a link to book some time with Sam. I think he has availability this week, so, yeah, we can go from there.
72 00:12:10.770 ⇒ 00:12:12.579 Mouhamad: Yeah, sounds good, sure.
73 00:12:12.580 ⇒ 00:12:16.250 Kaela Gallagher: Okay, cool, thanks so much for your time today, I appreciate it.
74 00:12:16.490 ⇒ 00:12:22.180 Mouhamad: Thank you so much for your time, thank you, have a good night. It’s not much for you, but it’s nice for us.
75 00:12:22.180 ⇒ 00:12:26.109 Kaela Gallagher: Yes, have a great rest of your night. Talk to you later.
76 00:12:26.110 ⇒ 00:12:28.880 Mouhamad: Have a great night. Talk to you later. Goodbye.
77 00:12:29.010 ⇒ 00:12:29.650 Kaela Gallagher: Bye.