The team you poach isn’t always the team you keep
Here’s a take you won’t hear much: Meta pulling seven founders away from Thinking Machines Lab isn’t a power move. It might be the clearest sign yet that a scrappy AI startup is doing something right — and that the biggest player in the room is nervous about it.
I’m Maya, and I write about AI for people who don’t have a computer science degree. So let me break down what’s actually happening between Meta and Thinking Machines Lab, and why this story matters even if you’ve never heard of either of them before today.
So Who Is Thinking Machines Lab?
Thinking Machines Lab — often shortened to TML — is an AI startup that has been quietly building something serious. Serious enough that Meta reportedly held talks to acquire the company around early 2025. That deal never happened. What happened instead was a slow, steady drain of TML’s founding team, with Meta picking off seven of its founders one by one.
On the surface, that sounds like a win for Meta. Big company, big salaries, big resources. Of course people leave for that. But zoom out a little and the picture looks different.
Why Losing Founders Isn’t Always Losing
When a startup loses its founders to a tech giant, the usual story is: startup dies, big company wins. But TML didn’t fold. Instead, it kept recruiting — pulling in researchers directly from Meta. The talent is flowing in both directions, and TML is still standing, still building, and still attracting serious people.
In December 2025, TML made a notable move by appointing Soumith Chintala as its CTO. If that name sounds familiar, it should — Chintala is a well-known figure in the AI research world. That appointment, which took effect in early 2026, sent a clear signal: TML isn’t a company in retreat. It’s a company on the move.
Think of it this way. If someone kept trying to buy your lemonade stand, and when you said no they started hiring your friends away, you’d probably feel pretty good about your lemonade recipe. Meta’s repeated attempts to absorb or dismantle TML — first through acquisition talks, then through talent poaching — read less like dominance and more like a company that can’t stop thinking about a competitor it publicly doesn’t need to worry about.
The Talent War Nobody Talks About Honestly
The AI talent war gets covered like a sports draft. Who got picked? Who got traded? Which team has the best roster? But that framing misses something important for regular people trying to understand where AI is actually heading.
The most interesting AI work right now isn’t always happening at the biggest companies. It’s happening at places small enough to move fast, take risks, and build culture around a specific vision. When a startup can lose seven founders and still recruit aggressively from the company that poached them, that tells you something about the strength of the idea at the center of it all.
TML is actively recruiting researchers from Meta even as Meta continues pulling from TML’s founding team. That’s not a one-sided talent drain. That’s a genuine tug-of-war between two organizations that both believe the other has something worth having.
What This Means If You’re Not a Tech Insider
You don’t need to follow every hire and departure to understand the bigger picture here. What this story really shows is that the AI space is not a settled hierarchy. The idea that a handful of giant companies have already locked up all the best minds and all the best ideas is simply not true.
Startups like TML are proof that talent, vision, and the right environment can compete with billion-dollar budgets. And when a company like Meta feels the need to repeatedly chip away at a startup rather than just outbuild it, that’s a signal worth paying attention to.
The founders who left TML for Meta may do great work there. The researchers TML is pulling back from Meta may do great work at TML. What matters is that the competition is real, the ideas are flowing in multiple directions, and no single company has a monopoly on what comes next in AI.
Keep Watching TML
If you’re new to following AI and you want one company to keep an eye on, Thinking Machines Lab is a solid choice. Not because it’s guaranteed to win anything, but because the fact that Meta can’t stop thinking about it is probably the most honest endorsement a startup could ask for.
Sometimes the best sign that you’re onto something is who’s trying to take it from you.
🕒 Published:
Related Articles
- Frozen at the Starting Line — Why AI and Humans Both Struggle to Take the First Step
- Überprüfen die UC-Universitäten die KI? Ihr Leitfaden zur Erkennung von KI & den Zulassungen
- 5 errori di monitoraggio dei costi che costano soldi veri
- Your AI Startup’s Shot at $100K Equity-Free Disappears Tonight