Australia, 6:47 AM. My laptop is heating up on my lap, my model has been refusing to converge for 20 minutes, and I'm sitting on my terrace in shorts and flip-flops.

A few years ago, white lab coat, climate-controlled lab at 18°C, I was pipetting DNA samples following a precise protocol. Same concentration. Same rigor. Just a completely different universe.

How did I get here?

It wasn't a "career change" as people imagine it. It was a logical evolution.

Before We Continue: A Bit of Context About My Relationship with Code

A bit of context about my relationship with code. Before I tell you how ML changed everything, I need to talk to you about something important.

In high school, I was the guy who avoided the computer science option.

Coding wasn't for me. Too abstract. Too cryptic. I felt lost from the first line. Others seemed to understand instantly, while I stared at the screen wondering if I just wasn't built for it.

Result? I turned to what spoke to me: biology. Living things. Stuff you can see under a microscope.

Spoiler: I was completely wrong.

Today, I code every day. And I love it.

What changed? I discovered that code is just a tool, not an end in itself. When you use it to solve problems you're passionate about, everything becomes clear. It's not a question of "innate talent" or "being wired differently."

It's just about finding the right angle of attack.

And for me, that angle was machine learning applied to biology.

The Thing They Never Tell You About Biology

When you're a biologist, what do you spend 80% of your time doing?

Analyzing data.

Experiment results, growth curves, endless statistics... It's pattern matching all the way. You search for signals in noise. You test hypotheses. You fail. You start over.

Exactly like in machine learning (ML), basically artificial intelligence.

The only difference? In bio, you wait 3 weeks for your cells to grow. In ML, you rerun your script and get your results in 3 minutes.

The day I understood that, everything changed.

And it was precisely this realization that pushed me to explore other horizons...

How ML Pulled Me Away from the Bench

My ML breakthrough? It came down to a single image.

I was reading a bioinformatics research paper during my master's. Researchers had used ML to predict protein interactions. Something that would have taken us months of experiments in the lab, with hundreds of manipulations...

Solved in a few hours of computation.

I had this game-changing thought: "Holy shit, but I could do this on my own projects and make money."

The following days, I dove into the algorithms. I struggled. I understood nothing for hours. Then something clicked. A few months later, I was running my first classification model on biological data.

I was hooked.

It was like going from walking to teleportation. All this computing power, this ability to test hypotheses in loops, to iterate in minutes instead of weeks...

And the craziest part? My scientific skills became a superpower.

I understood the data. I knew which features made sense. I spotted potential biases at a glance. I spoke the language of both worlds.

Suddenly, I wasn't just a biologist learning to code. I was the bridge between two universes that no one really connected.

But beyond the technical excitement, there was something deeper pushing me toward the exit...

The Real Reason for My Departure: Toxic Lab Culture

They sell you research as something noble. And it's true, on paper it's magnificent.

They also sell you that after your PhD, you become a researcher. That's a lie.

The reality? There's an endless obstacle course. Competitions. Repeated fixed-term contracts. Years of uncertainty before maybe landing a permanent position.

But that's not the worst part.

It's the atmosphere.

The hierarchy that treats you like a lackey. Lab heads who treat interns like servants who fetch coffee. Team meetings that are supposed to be scientific but turn into pissing contests between bosses wanting to prove who's the best. Jealousies between teams.

You spend your time navigating politics instead of doing science.

And the waiting. Damn, the constant waiting.

Waiting for your cells to grow. Waiting for equipment to be available. Waiting for peer-reviews that take 8 months. Waiting for a research director to deign to look you in the eye.

In ML? If you have an idea at 11 PM, you test it before midnight. Doesn't work? You pivot immediately. Works? You deploy.

It's the same scientific rigor, but in fast-forward mode. And most importantly, without the political bullshit.

I loved research. But I was tired of wasting my time in a system that crushes people instead of helping them grow.

Australia: The Springboard to a New Professional Culture

So why Australia?

Not for the kangaroos or beaches (although that's a nice bonus).

It was a first step toward independence. And above all, an opportunity to discover a radically different professional culture.

Here, no one treats you like a nobody because you don't have 20 years of seniority. The hierarchy exists, but it doesn't crush you. People respect you for your skills, not for your title or affiliated lab.

And most importantly, the Australian tech ecosystem is insane. Startups everywhere, remote-first culture, accessible international clients, and a "let's try and see" mentality instead of "but who's going to validate this project for 3 years?"

It was exactly the shock I needed after years in French labs.

And now that I've tasted this freedom, I want to go even further...

Next Level: Total Independence

Because now that I've seen another way of working exists, I want to build my own.

More varied clients. More projects I'm passionate about. More control over my schedule. No more toxic hierarchy.

The freelance/independent IT status here is an opportunity machine. You can work for companies in Australia, Europe, Asia... as long as you have your laptop and wifi.

Australia was the springboard. Independence is the destination.

And that's exactly what I'm building now.

What No One Tells You About This Kind of Journey

People will tell you:

"You did all those years of study for nothing"
"You're starting from scratch"
"It's risky" - welcome to Europe

The reality:

My years in biology are my biggest asset
I'm not junior, I just changed tools
The only risk was staying stuck in a toxic system

If you're a scientist wondering if ML is for you, let me tell you something:

You already have the hardest skills.

Logic, rigor, problem-solving, perseverance in the face of repeated failures... you have all that.

Learning Python, NumPy, TensorFlow? That's just syntax. You can learn it in a few months.

And now, what am I going to do with all this?

What's Next

This blog is my space to share this journey but also my knowledge through many original articles.

I'm going to talk about:

How science and ML complement each other
The concrete projects I'm building
The freelance adventure in Australia
Mistakes to avoid (I've made plenty)
Resources that really helped me

Because this kind of journey isn't talked about enough. And yet, more and more of us are doing it.

If you're also a scientist looking at ML wondering "why not?", or if you're struggling in a toxic lab telling yourself there must be something else...

Stick around.

We're going to have fun.

Frederic Navez

Converted biologist. ML Engineer. Soon-to-be freelancer. Always curious.