Senior Research Scientist/Engineer, Deception

The role

You will lead research projects on white-box deception detection, probe-based monitoring, and training against probes. This involves running post-training and RL experiments on large models and designing model organisms to stress-test methods.

What you will do

• Lead research projects on white-box deception detection, probe-based monitoring, and training against probes.
• Define research questions, set milestones, choose experiments, and assess results.
• Run post-training and RL experiments on large models.
• Diagnose training instability, reward hacking, and infrastructure failures.
• Design model organisms that stress-test methods against realistic failure cases.
• Mentor Members of Technical Staff.
• Partner with the infrastructure team to plan and run large post-training jobs.
• Enable frontier labs to adopt methods through reproducible code and collaboration.
• Write and present findings for research papers and the AI safety community.

Required qualifications

• 3+ years experience with implementing post-training, model training, or closely related efforts in a frontier AI environment
• Hands-on experience post-training models at scale (RL, RLHF/RLVR or large fine-tuning runs), ideally on models of 400B+ parameters
• Strong Python and PyTorch
• Experience working in large, shared research codebases
• Strong research judgement
• Based in or willing to work in U.S. time zones

Nice to have

• Safety or alignment research at a frontier lab or a research lab working on large models (1T+ parameters)
• 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
• A PhD in a relevant field

Before you apply

• 1 short intake call with a technical recruiter
• 1 30 minute screening call with the Head of Engineering
• 60 min ML assessment
• 60 min code review interview
• team fit assessment with the hiring manager

About the company

FAR.AI is a non-profit AI research institute working to ensure advanced AI is safe and beneficial for everyone. The institute facilitates breakthrough AI safety research and advances global understanding of AI risks and solutions.