Lead Researcher, Pre-Training Safety & Safe AI

Aimlroles; Berkeley (CA); Hybrid; USD 290,000 - 450,000; Full time

Benefits offered by this job

Catered lunch and dinner at Berkeley

Job summary

FAR.AI is hiring a Research Lead to develop and lead our work on pre-training safety, shaping models’ capabilities at their source. You’ll guide the research agenda, collaborate with our red team to stress-test frontier models, and scale methods like data filtering and unlearning.

This role offers high autonomy in an impact-driven environment with strong focus on safety and governance. You’ll build and lead the team, mentor staff, and stay hands-on to write code and run experiments.

Qualifications

• Strong existing research track record in AI or related field.
• Experience with language-model pretraining and dataset construction.
• Ability to build and debug research systems and pipelines.

Responsibilities

• Articulate a research agenda with theory of change to mitigate risks from AI systems.
• Grow and lead a technical staff team.
• Lead novel research projects with unclear success markers.
• Publish findings and present at conferences and brief policymakers.

Skills

AI safety research

Pre-training safety

Language-model pretraining

Dataset construction

Controlled training experiments

Pipelines for scoring/filtering

Experimental judgment

Research leadership

Technical communication

Tools

Python

PyTorch

Distributed training

Job description

FAR.AI is hiring a Research Lead to develop and lead our work on pre-training safety, shaping models’ capabilities at their source. You’ll guide the research agenda, collaborate with our red team to stress-test frontier models, and scale methods like data filtering and unlearning.

This role offers high autonomy in an impact-driven environment with strong focus on safety and governance. You’ll build and lead the team, mentor staff, and stay hands-on to write code and run experiments.