jobs
Research Engineer, AI Monitoring
About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole.
En inglés
- Company
- Anthropic
- Location
- London, UK
- Status
- Open
- Posted
- 2026-11-16 21:28 UTC
Anthropic's AI monitoring team builds and runs classifiers that stop internal agents from taking dangerous actions, such as sandbox escapes or data exfiltration, blocking a flagged action before the tool call runs. The research engineer designs, trains and evaluates those classifiers, deploys them into RL training, evaluations and internal agents, and owns their accuracy, latency and cost. The role covers datasets, labelling pipelines and evaluations, and stress-testing with red teams. The posting asks for prioritisation across parallel workstreams and critical review of code and experiments, including work produced by AI agents.
Original job posting