AI Behavior Engineer
Transluce is an independent San Francisco 501(c)(3) nonprofit research lab building open technology and research infrastructure for scalable oversight and understanding of AI systems.
Funding history
About Transluce
Transluce develops research, platforms, and open-source tools intended to help evaluators and other stakeholders understand, measure, and steer advanced AI behavior in the public interest. Its active product, Docent, analyzes AI-agent transcripts using traceable behavior rubrics, qualitative review, and quantitative analysis.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will design, prototype, and run behavioral AI evaluations for emerging policy and oversight needs. You will deliver evaluations for government and civil-society partners, conduct privileged-access oversight exercises with frontier labs, and adapt evaluation pipelines with domain experts.
Requirements
- Experience designing and running AI evaluations
- Experience with behavioral, interactive, multi-turn, agentic, or red-teaming evaluations
- Strong engineering judgment
- Customer-facing, consulting, or forward-deployed experience
- Experience running evaluations at scale or in production
- Ability to balance researcher, domain-expert, and decision-maker needs
- Strong communication and feedback skills
Responsibilities
- Scope, prototype, and run behavioral AI evaluations
- Respond to emerging policy and oversight needs
- Execute evaluation contracts with government evaluators
- Build evaluations for harmful manipulation
- Design and run privileged-access evaluations
- Conduct external oversight exercises with frontier labs
- Adapt behavioral evaluation pipelines with civil society organizations and domain experts