Research Scientist

FAR.AI is seeking Research Scientists to work on impactful AI alignment and security projects, focusing on scoping research directions, implementing and running experiments, and analyzing and writing-up results.

FAR.AI is a non‑profit AI research institute dedicated to ensuring advanced AI is safe and beneficial for everyone. Founded in July 2022, the institute has grown to over 40 staff and publishes work at top conferences such as NeurIPS, ICLR, and ICML.

We conduct cutting‑edge AI safety research in-house and support the broader community through grants. Our approach includes rapidly scaling research that shows high potential for impact and demonstrating viable prototypes to AI companies and governments.

As a member of our technical staff, you will take ownership of and accelerate existing AI alignment research agendas. You will collaborate closely with engineers, receive mentorship through code review and pair programming, and develop scalable implementations of machine learning algorithms for scientific experiments. You are encouraged to propose novel directions, join a research pod aligned with your interests, and have the opportunity to attend conferences free of charge.

• Scope and prioritize research directions in AI alignment and safety.
• Design, implement, and run experiments on frontier models.
• Analyze results and write up findings for publication.
• Collaborate with engineering teams to build scalable tooling and infrastructure.
• Participate in research pods and contribute to collective scientific progress.

• PhD or equivalent research experience in computer science, AI, machine learning, or statistics.
• Experience leading or playing a senior role in research projects related to machine learning.
• Ability to communicate novel methods and outcomes to both technical and non‑technical audiences.
• Unconventional or under‑explored AI alignment ideas.
• Interest or experience in areas such as evaluation and red‑teaming, infrastructure, deception mitigation, adversarial robustness, or mechanistic interpretability.

Position is available remote (global) or in person in Berkeley, CA. The salary range is $120,000–$190,000 per year, depending on experience and location. We also cover work‑related travel, equipment expenses, and offer catered lunch and dinner at our Berkeley office.