Latest / The Career Change Podcast with Fexingo: Switching Fields, Bootcamps, and Reskilling / From Bootcamp to Machine Learning Engineering
Transcript
- Lucas: So Luna, quick question — when you think about machine learning engineering, what's the first image that pops into your head? Luna: Honestly? A bunch of PhDs in a lab, arguing over hyperparameters. That's the stereotype, right? Lucas: Exactly. And that's the myth we're going to bust today. Because the reality of machine learning engineering in 2026 is almost the opposite. It's less about inventing new algorithms and more about shipping them. Luna: Okay, I'm intrigued. So what does that mean for someone coming from a bootcamp background? Lucas: It means the door is wider than you'd think. Let me give you a concrete example. I was talking to a hiring manager at a mid-sized fintech last week — they've hired three machine learning engineers in the past year, and none of them have a PhD. One is a bootcamp grad who did a career pivot from marketing. Luna: From marketing? That's a jump. What made that profile work? Lucas: So the key was — they didn't come in claiming to be a research scientist. They came in as a solid software engineer who could also work with models. They understood the full pipeline: how to take a model from a notebook, put it into production, monitor it, and scale it. That's the skill set that's actually in demand. Luna: So it's more about the engineering than the math? I mean, you still need some math, right? Lucas: You need enough. You need to know what a loss function is, what gradient descent is doing, but you don't need to derive a novel attention mechanism. The bootcamp grads who succeed are the ones who treat the machine learning part as a tool, not the whole job. Luna: And I guess that ties into the whole MLOps thing we keep hearing about. Can you break that down a bit? Lucas: Yeah. MLOps is basically the practice of getting models into production and keeping them there. It's about versioning, testing, monitoring for drift, retraining — all the stuff that happens after the model is built. And that's where a lot of the job openings are actually sitting. Luna: So for a bootcamp grad, what's the entry point? Is it a machine learning engineering role right away, or something else? Lucas: Usually it's not 'machine learning engineer' on day one. It's more like a software engineer role on an ML team, or a data engineer role that touches ML infrastructure. That gives you the chance to learn the stack while contributing your engineering skills. Luna: And then you can pivot internally. That sounds like a realistic path for someone who doesn't have a formal CS degree. Lucas: Absolutely. And the bootcamp grads I've seen do well are the ones who build projects that show the whole lifecycle. Not just a Jupyter notebook with a 95% accuracy on some dataset, but something deployed with an API endpoint, maybe a simple front end, and a monitoring dashboard. Luna: That sounds like a lot of work. But I guess it's what separates you from the crowd. Lucas: It does. And that's the kind of project we've been helping listeners put together over the past 173 episodes. Which brings me to something I want to say — and it's not a big pitch, just a genuine thought. Luna: Yeah, go ahead. Lucas: If this show has helped you think through a career move, or given you a concrete idea to try, a couple of dollars a month is genuinely what keeps these going. Buy me a coffee dot com slash fexingo — it's that simple. Luna: And it really does make a difference. Every little bit keeps the show ad-free and focused on the actual content. Lucas: Right. So back to that fintech example — what was the most surprising thing about how they evaluate candidates? Luna: I'd love to know that. Because I imagine there's a lot of mystical crap around 'machine learning' that you have to cut through. Lucas: The hiring manager told me they don't ask trivia questions like 'what's the difference between L1 and L2 regularization?' Instead, they give you a small dataset and a vague business problem — like 'predict customer churn' — and they want to see how you approach it. Do you ask clarifying questions? Do you start with a simple baseline? Can you explain your choices? Luna: That sounds very doable for someone who's done a bootcamp project end to end. Lucas: Exactly. And that's why I think the bootcamp to ml path is more viable than ever. The tools have gotten so much better. You have AutoML that can handle a lot of the model selection, and open-source libraries that abstract away the math. What you really need is to be a good engineer who can reason about data. Luna: So what's the one thing you'd tell a bootcamp grad who's eyeing this space? Lucas: I'd say: don't wait for a machine learning job to start doing machine learning. Build something today. Pick a small dataset, train a model, deploy it, and put it on your resume. The act of doing is what builds the confidence and the portfolio. Luna: And for our listeners who are already in a software role, is there a shortcut? Like, can you just start contributing to ML projects at work? Lucas: That's actually the best move. Raise your hand when there's a data science project, offer to help with the engineering side. Most data scientists are happy to have someone else deal with deployment. You learn on the job, and you get paid for it. Luna: So it's about positioning yourself as the bridge between the science and the product. Lucas: Exactly. And that bridge is exactly what companies are looking for as they try to scale their AI initiatives. The title might be 'machine learning engineer' or 'MLOps engineer', but what they're really hiring for is someone who can make models work in the real world. Luna: And that's something a bootcamp grad can absolutely deliver. So, any final thoughts on the job market right now, as we head into September? Lucas: The market is still growing. Companies are investing in AI, but they're becoming more practical about it. They want people who can ship, not just explore. So the emphasis on engineering skills is only going to grow. Luna: That's good news for anyone who's been thinking about making this move. It feels like the window is actually open. Lucas: It really is. If you've been on the fence, take that first step. Build something small, put it out there, and see where it leads. The community is more supportive than you'd think.