Member of Technical Staff (Post Training)
Inherent · London
Frontier labonsitesenior
At a glance
Highlights
- Train AI scientists that improve the very pipeline that trains them.
- Build autocurricula turning open-ended research tasks into reward signal.
- Shape core research of a frontier lab building AI for scientific discovery.
Ideal candidate
A deeply experimental deep learning researcher who is AI-pilled and thrives in high-intensity, in-person collaboration. You care about recursive self-improvement and building AI scientists that accelerate their own creation.
Requirements
- 3+ years of deep learning research experience
- Experience post-training large language, vision, video or multi-modal models
- Demonstrated track record in deep learning research via papers, model releases, open-source contributions or other artifacts
- 5+ years of software engineering experience including Python and PyTorch or JAX
- Experience using the latest coding agents with opinions on optimal workflow
- Enthusiasm for experimental organizational design
- AI-pilled: adopting agents and keen to build a company where agents are front and centre
- Hands-on experience training LLMs with RL at scale preferred
- Familiarity with distributed and long-context training infrastructure preferred
- Background in autocurricula, open-endedness, meta-learning or recursive self-improvement preferred
- remote policy
- in-person every day
Lab profile
About Inherent
Frontier lab building recursively self-improving AI to discover new scientific knowledge.
Known for
- Recursive self-improvement
- Post-training
- AI for science
- Training and inference systems
Key facts
- Based in King's Cross, London
Posted · Last seen