Model Adaptation — Member of Technical Staff
Berlin, Full-time, In person
Who are we:
Models should learn from what happens after deployment. We're building the loops that make this possible.
We will be the default inference provider for specialized tokens, running models that continuously adapt to each customer and workload. That means serving many continuously adapting models with frontier-level performance and attractive economics.
The work:
- Policy updates from live model interactions
- Turning agent traces into synthetic RL environments
- Neural KV-cache compaction and learned memory
- Designing rewards, evals, and training methods from scratch
This is open-ended research with a short path to production: read papers, question them, run experiments, understand weird failures, and ship what works.
No long post-training track record required. We're looking for creative, high-agency builders who have mastered something hard and enjoy problems where nobody knows the answer yet.
We work together in person from our office in Berlin.
Interview process
- Two technical interviews
- An onsite interview at our Berlin office
Apply
Tell us about exceptional work you've done.