Research Scientist/Engineer (Evaluations)
Apollo Research · London & San Francisco
CHECKED 7H AGO
Research instituteonsite
At a glance
Highlights
- Design automated pipelines assessing misalignment in frontier post-training.
- Red-team new model checkpoints with OpenAI, Anthropic, and DeepMind.
- Run pre-deployment evaluation campaigns every two weeks at scale.
Ideal candidate
A rigorous tester who loves probing frontier models for misalignment and building efficient pipelines to surface subtle behaviors from messy data.
Requirements
- Strong software engineering experience shipping production Python code
- Extract signal from large, messy datasets and spot anomalous behaviors
- Succinctly convey qualitative and quantitative findings to varied audiences
- Experienced AI power-user with different models and new workflows
Lab profile
About Apollo Research
AI safety organisation detecting and mitigating scheming in frontier AI systems and coding agents.
What they build
- Watcher – coding agent security monitoring product
Known for
- Pre-deployment testing with frontier AI labs
- Scheming detection evaluations
- Science of scheming and mitigations research
- Production monitoring of billions of agent tokens monthly