How to choose which projects to work on - Soma Froghyar

00:00

When I was looking for jobs, I was mainly looking at machine learning engineering, but without professional experience. And having had two years experience, I was pushed back into the full stack world.

00:13

Why are you a software engineer?

00:15

I started coding when I was 13, 14, I think. Honestly, the whole thing is kind of crazy. And my family just started watching a random series, and I just didn't want to watch that series, so I started coding because I wanted to learn C++. I was like, "You know what? Why do games work? What happens there? Why does my character jump when I press X, why does it jump? I want to know".

So I went on and did it. I was like, okay, C++ then I went to web development for like a few years as I was in high school years. I was just doing random websites for friends or this and that. And then I got into uni for software engineering because I knew I wanted to be building stuff continuously, because I was a hobbyist even before and I started doing a lot of data science related things. And I am very much into the machine learning engineering part of things.

So for my thesis that I sent you earlier, I'm not sure if you had a time to check it out. I made a [inaudible 00:01:24] recognizer system built with PyTorch with seven different models, different configurations and ways of regulating that convolution neural network. Aim is now to become a machine learning engineer, and build my own pipelines and everything because I'm a big fan of software architecture and just the whole microservice things and everything. I just love messing around with those. It's been a great journey as an engineer and I like where I am right now, honestly.

01:56

What's been the most meaningful project to you that you've worked on over the last two years or so?

02:03

I think the one that I'm expanding on right now. When COVID started, I isolated with a couple of my friends, just the four of us for like two, three months, barely seeing anyone else. When the whole thing started, I had a family vacation set up in at a [inaudible 00:02:28] for March or something, but that got canceled obviously.

So, that was one occasion when most of the family would've been together. And when we usually are together, we play this card game called [inaudible 00:02:44] it's a hand game and card game. We love that thing, whatever, but we didn't have the chance to meet and ultimately do that.

So my reaction to that was I'm an engineer, I'm going to build an online card game for this specific game that I've never seen online or didn't see it already implemented. So I built it with Socket.IO And Angular front-end. It worked for one single server that I got from Digital Ocean.

I made it kind of junkily, I'm not going to lie, but I just wanted to do it as fast as possible. Bear in mind, I was also full time working at a company as a software engineer with a lot of responsibilities there. So afterwards I was just working on this for like a month or two and it worked. We played a few times. It was very nice. They really appreciated it. They gave harsh feedbacks, which was just painful, but that's part of it.

03:47

That sounds amazing.

03:53

So I guess the fact that that worked out is one of the best feelings.

03:57

How have you gone about prioritizing what types of things you want to look for in order to in the longer term optimize for running your own thing?

04:08

I've been working on my thesis visualization, and now the card game, but slowly as I'm enjoying life, I guess.

04:18

Have you started looking for something else?

04:22

I'm very positive about the next steps. I'm going to be doing most likely software consultancy, and that would be talking with clients, help them with their ideas and do pretty much what I'd like at this point.

04:40

Sounds like in some ways that's another kind of extrapolation of you wanting to figure out being in a place where you can be exposed to multiple things that wants to accelerate your learning in some way.

04:57

Yeah.

04:58

So when it comes to figuring out what you do next, and what company you want to work in next.

05:07

It's important for me to have a good project that I can work on or that is going to be cool. But now I'm also realizing the past months that I do like to build things and then do it the right way. So when it comes to this idea of software consultancy in this sense, I like to put a good start for anything or either a good feature that is very well done, and then I'm sure that this is going to work for them for good while. I think that would be satisfying to me as a person anyway.

So when I was looking for jobs, I was mainly looking at machine learning engineering, but without professional experience and having had two years experience, I was pushed back into the full stack world, but I do want to make the switch to machine learning engineering, and that was also what I tell recruiters whenever I was talking about their prospects, what they do want from us. And yeah, I just laid down that I want to be a machine learning engineer down the line, but then again, I do like full stack obviously because it's a passion of mine.

06:28

And in the long term, what do you think doing your best work will mean to you?

06:35

If it's all mine in a way or started by me completely and is helping people, and does have a good impact on the team, on the engineers. But then again, as in like, if I build a team or like some open source project, it would be something that is meaningful. Some people, whether it be any library that might make someone's life two times easier for like a week, that'd be great. I don't mind doing that. Right now, I just want to develop myself and put out good stuff like quality software, like with this card game, I'm going to try to do it the best possible way.

07:18

How have you gone about choosing which companies you talked to?

07:24

I was never really looking, but people came to me because of the two years experience. So I made an account on cord as well, obviously, and I got random offers, conversations, going some interview processes, and I took them as I want to listen to what you guys are about, like what the project is. I'm explaining who I am in those terms, and they are interested in moving me forward, then we do the interviews. Like as an engineer for two years, I was up for a senior position. If they go too technical, I'm kind of upset about it, but slowly I'm going to move into machine learning engineering so I know that I have a good place there too rather than opt to be a full stack engineer.

08:09

When you're in a position where you can take a pick and decide who you want to talk to, what are the factors that you take into account?

08:23

The project itself. If it's something that I can kind of... As you say, as I optimize every little conversation I have, because I know engineers who work full time and then are looking for other engineers, I've been there. It's rather explain me what we're going to do if I join. Because I know you're an engineer, I'm an engineer. Let's discuss what's going on because that's easier to get straight into it because then if I know that it's not something I want, then they shouldn't have me because my heart is not going to be completely in it.

So it's rather what I can build with them in terms of what the product is. Does it mean enough for me or is it sustainable or stuff like that? Like these factors do add something to what I ultimately want with something that is perfect for my development, which is optimization, which I'm doing in my life most of the time.

09:36

Get to the heart of what the actual problems are that people are facing.

09:43

What do you mean the actual problems?

09:47

So I guess there's like the company wide problems, but then there's also the engineering challenges that you're not going to get from a recruiter or necessarily from a company founder.

09:59

I assume you do. You do if you ask.

09:59

How do you get to those problems?

10:02

I do ask, and they are usually open about it because they work there, they know what they need. I know if they specifically need someone real fast and they usually have tech debt needs work working, which is fair enough. I had to deal with a lot of tech debt, but I know it's not fun and I don't know what they need the engineers for. So I ask like, okay, so when I'm coming in, what would we be working on? Or what is the project that is requiring more engineers?

And sometimes as it's tech debt, and they tell me that, and they're very honest about it. I'm like, thank you. That's all I need to hear. That's perfect, fine. I just know what is there, but I still want to hear about the project because if the tech debt goes away, then I can't be there for that because I'm not in a hurry to join right now.

Otherwise, I'm very open in every single conversation just to show that I'm not a bullshitter, I guess. Let's build something. Just tell me what to do. Tell me what's the project. I can handle a lot of stakeholders. Sure.

11:17

I think this is answer is different for everyone, but I'd love to know, I guess, if you were to try and summarize it as succinctly as you possibly could, what does doing your best work mean to you?

11:37

It's building something meaningful very well and while not sacrificing on your own happiness. I would say that's a good job.

11:50

Love it. That's really interesting.

11:52

That's a good job. If you do that, that's a good job in my opinion.

12:01

I hope that we can all do that in some way.

12:05

Sure.

12:06

And then my last question is, what advice would you give to someone who's trying to figure out where they want to work?

12:14

Be picky. Why not? You're an engineer. If we're talking to engineers right now, just be picky to whatever you want to do, because it might be just some legacy thing that you might love and then you might actually get better pay for. So just do that. Or if you hate, then don't for the sake of it. Just be better at what you enjoy.

12:37

Love it. Thanks so much taking the time out, Soma.