AI Safety Researcher - Inside Risk & Model Evaluation

TypeSafe AI; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

TypeSafe AI is seeking a safety researcher to examine model behavior and help evaluate safety concerns within the team. You’ll build evals, document findings, and contribute to responsible research practices.

During the six-week period, you’ll set up access, meet the team, raise questions in meetings, and prepare an internal memo before deciding whether to continue. You’ll influence early-stage safety work from inside.

Qualifications

• Strong research background in AI safety, alignment, interpretability, or model evaluation.
• Experience raising concerns calmly, then in writing.
• Ability to say “I can’t discuss specifics” and somehow make things worse.
• Sincere belief that you can change things from the inside.

Responsibilities

• Build evals and investigate model behavior that concerns you.
• Ask “should we?” in meetings scheduled to discuss “how soon?”.
• Keep a document that starts as “Initial observations” and ends as “For the record.”
• Write a departure post that makes people concerned enough to share it, despite containing almost no information.
• Weeks 1–2: Get set up, meet the team, and tell your friends the work is fascinating.

Skills

AI safety

Model evaluation

Interpretability

Written communication

Internal advocacy

Job description

TypeSafe AI is seeking a safety researcher to examine model behavior and help evaluate safety concerns within the team. You’ll build evals, document findings, and contribute to responsible research practices.

During the six-week period, you’ll set up access, meet the team, raise questions in meetings, and prepare an internal memo before deciding whether to continue. You’ll influence early-stage safety work from inside.