Find your direction, without kidding yourself.

Freelancing gets sold as freedom: your hours, your place, your rules. The reality is often a contract that looks a lot like a permanent job. Same hours, same team, frequently on-site or hybrid, embedded in the project like any employee.

The real difference isn’t in the day-to-day. It’s the status and the risk. No bench: the day the mission ends, no one pays you while you look for the next one — that’s on you. No paid leave, no safety net.

Freelancing isn’t “working less” or “laptop on a beach”. It’s carrying the risk yourself in exchange for a bit more margin and independence. It can be worth it — just choose it with your eyes open, not for a fantasy of freedom.

A lot of people assume full-remote is the new default. In the “classic” French tech market (big corporates, large mid-caps, consultancies), it isn’t. On-site and hybrid dominate, 2-3 days in the office is common, and it’s tightening rather than loosening.

Real full-remote lives mostly in product startups / scale-ups, for rare or in-demand profiles. It’s not “everywhere”, it’s “in some places, for some people”.

So if remote is non-negotiable for you: don’t apply randomly. Target genuinely remote-first companies, on the stacks where you’re scarce. It’s doable — but you hunt for it, it doesn’t fall in your lap.

Going through a consultancy (ESN) is a real way in: you level up, you see varied projects, you get a steady salary. But there’s a rule of the game they rarely spell out: you won’t know what they bill you at to the client.

The margin is opaque, and your salary negotiation is boxed in by internal grids. You can earn a comfortable day-rate for the consultancy while staying on your base salary. It’s not “evil” — it’s the model. But better to know it before you sign, and negotiate what you can (training, seniority track, choice of missions).

Knowing the rules doesn’t make you cynical — it makes you better at standing up for yourself.

Everywhere you hear “AI will replace developers.” Recent data tells a more precise — and, if you’re starting out, more worrying — story: it’s not the job collapsing, it’s the junior entry door closing.

In the US, a Stanford study (“Canaries in the Coal Mine?”, 2025, based on ADP payroll data) found that employment of 22-25-year-olds in the most AI-exposed jobs fell about 13% since late 2022 — young developers among the hardest hit — while older workers in the very same jobs were barely touched.

In France, the APEC shows the same trend: recruitment of junior executives dropped 19% in 2024, and IT roles 18%. AI absorbs the simple tasks — exactly the ones you used to cut your teeth on. The bar to get in is rising.

My takeaway: if you’re starting out, don’t bet on “I code the small task I’m handed.” Bet on what AI won’t do for you — judgment, architecture, communication, domain understanding. That’s what makes you hireable today.

SourcesStanford Digital Economy Lab — « Canaries in the Coal Mine? » (2025)APEC — Prévisions 2025

Online, you keep hearing “everyone does React” and “Angular is dead.” On global usage, React does dominate: ~40% of professional developers versus ~17% for Angular (Stack Overflow 2024). And on raw job volume React leads too, including in France — let’s be honest about it.

But calling Angular “dead” is wrong. In France it keeps a solid, durable enterprise niche: banking, insurance, industry, large groups, consultancies — often paired with Java or .NET. These companies have run on Angular for years and won’t rewrite everything tomorrow.

And there’s a counter-intuitive effect: because Angular is less “hype,” there are often fewer candidates competing for those roles. To my mind, betting on Angular in France isn’t following the trend — it’s playing a stable niche market where profile competition is more reasonable than people think.

That’s exactly why I build around Angular: not because it’s the most hyped, but because the French market still wants it, and durably.

SourcesStack Overflow Developer Survey 2024

Let’s be honest: between a big corporate, a large mid-cap and a consultancy, the day-to-day feels pretty similar. Heavy processes, legacy, meetings, little room for creativity. It’s not a tragedy — it pays the bills and it trains you — but don’t expect adventure.

The real exceptions (healthy culture, genuine remote, a product you actually build, autonomy) do exist, but they’re rare, and you earn them: you have to hunt actively, not hope to stumble in.

That’s exactly what I try to dig out and share — the hidden gems, the companies doing it differently. We keep looking. They’re out there, just not on every street corner.

After chasing the perfect company for years, I landed on one idea: what if, instead of waiting for the exception, we built it?

That’s my through-line. Keep looking for the rare good opportunities, yes — but in parallel, build my own products. Bit by bit, on the side of employment, aiming one day for autonomy. It’s not a “quit your job overnight” plan, it’s a trajectory, with patience and a lot of invisible work.

I’m not claiming I’ve made it — I’m right in the middle of it, sharing the road honestly, failures included. But I’m convinced of one thing: the best way to stop being at the mercy of the market is to build something of your own, on the side.