Senior Research Scientist/Engineer, Deception

Highlights

• Lead research scaling white-box probes for deception detection.
• Turn probes into real-time monitors for deployed frontier models.
• Test if honesty training makes models honest or better at hiding.

Ideal candidate

Experienced research scientist/engineer passionate about AI safety who enjoys leading projects end-to-end and scaling honesty and deception detection methods for real-world adoption.

Requirements

• Experience working in large, shared research codebases
• Strong research judgement to know what experiments to run next and when a result needs scrutiny
• Based in or willing to work in U.S. time zones
• Safety or alignment research at a frontier lab or with large models
• Experience with interpretability or white-box methods (e.g. probes, steering or SAEs)
• Experience building model organisms of misalignment, or studying reward hacking, scheming or evaluation awareness
• Publications at top ML venues (NeurIPS, ICML, ICLR) or widely cited technical reports
• PhD in a relevant field