Senior AI Safety Research Scientist - Deception & Monitoring

Far Ai; United States; Hybrid; USD 225,000 - 350,000; Full time

Job summary

FAR.AI, a non-profit AI safety research institute, seeks an experienced research scientist/engineer to partner with the Deception team's Research Lead. You’ll drive research projects from proposal to implementation across multiple projects and teams.

You'll work on deception detection, real-time monitoring, and training for honesty, including post-training experiments on large models (400B+ params), RL/RLHF, and scalable infrastructure. Berkeley in-person preferred.

Qualifications

• 3+ years of experience 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, and experience working in large, shared research codebases.
• Strong research judgement to know what experiments to run next and to recognize when a result needs further scrutiny.
• Based in or willing to work in U.S. time zones. In person in Berkeley is preferred.

Responsibilities

• Lead research projects on white-box deception detection, probe-based monitoring, and training against probes.
• Run post-training and RL experiments on large models, diagnose training instability and reward hacking.
• Design model organisms that stress-test methods against realistic failure cases.
• Mentor Members of Technical Staff to strengthen research judgment and experimental rigor.
• Partner with the infrastructure team to plan and run large post-training jobs on FAR.AI’s compute cluster.
• Enable frontier labs to adopt the team’s methods through reproducible code and evaluations.
• Write and present findings for research papers, frontier lab partners, and the broader AI safety community.

Skills

Post-training modeling

RL/RLHF experience

Python

PyTorch

Large-scale experiments

Research judgment

Education

PhD in a relevant field

Tools

Compute clusters

RL frameworks

PyTorch

Job description

FAR.AI, a non-profit AI safety research institute, seeks an experienced research scientist/engineer to partner with the Deception team's Research Lead. You’ll drive research projects from proposal to implementation across multiple projects and teams.

You'll work on deception detection, real-time monitoring, and training for honesty, including post-training experiments on large models (400B+ params), RL/RLHF, and scalable infrastructure. Berkeley in-person preferred.