Technical Program Manager - Embedded Evaluations and Investigations
As Technical Program Manager at Transluce, a nonprofit AI-evaluation research lab, you scope, organize, and run end-to-end evaluation engagements for frontier AI labs, governments, and civil society partners — coordinating technical staff, owning partner relationships, and communicating results. The role is in-person in San Francisco with regular travel.
- Compensation
- $250K–$350K / yr
- Location
- San Francisco, USA
- Work setup
- Not specified
- Seniority
- Not specified
- Employment
- Not specified
- Experience
- Not specified
- Category
- Embedded
- Posted
Skills
Must have
- Understanding of AI evaluations and ability to engage with technical teams on their design and results
- Track record of managing cross-functional teams (tech company, consulting firm, or fast-paced startup)
- Experience engaging external partners such as frontier AI companies, government actors, or civil society organizations
- Ability to work through ambiguity and deliver against complex, evolving goals on short timelines, building workflows where none exist
- Ability to translate technical research for a range of audiences
- Exceptional communication and relationship skills, low ego, openness to giving and receiving feedback
- Willingness to work in person in San Francisco with regular travel
Nice to have
- Experience managing evaluations at scale or in a production context
- Proficiency in Python and data analysis tools (e.g., pandas, SQL)
- Familiarity with common ML and evaluation frameworks (e.g., PyTorch, Hugging Face, Inspect)
- Experience building and applying software tools, for example in a forward-deployed engineering role
- Knowledge of frontier AI risk management practices or relevant regulatory guidance
Full description
Salary range: $250,000 - $350,000/year + benefits + bonus
Description: Transluce is a fast-moving nonprofit research lab building the public tech stack for AI evaluation and oversight. We partner with frontier AI developers, governments, and civil society organizations to evaluate how AI systems behave, from misaligned agent swarms to child safety and mental health.
About the role: As a Technical Program Manager, you will scope, organize, and direct the work of our technical staff across our external engagements. These include embedded evaluations on site at frontier labs, rapid-response incident analysis, and evaluation services for partners such as the EU AI Office and Common Sense Media. You will own each engagement end to end, from the first scoping conversation to the final dissemination.
As an early member of a highly collaborative team, you will learn and grow quickly, and work closely with our technical and governance teams to protect researcher time and turn one-off engagements into repeatable playbooks.
Core responsibilities:
- Define the scope of each engagement, set key milestones, and negotiate terms, with support from our operations team.
- Organize the technical team's work, manage dependencies, and keep everyone aligned through frequent touchpoints.
- Act as a sounding board on key strategic and methodological questions, reflecting the needs of the engagement and other considerations, such as relevant regulatory guidance.
- Own the day-to-day partner relationship, building trust and resolving issues while protecting Transluce's core interests and the technical team's time.
- Engage directly with evaluation results and technical analysis to understand the work, identify issues, and help drive the engagement forward.
- Present deliverables to partners and summarize results for broader audiences, including blog posts, papers, and government and media briefings.
Minimum qualifications:
- Understanding of AI evaluations and the ability to engage with technical teams on their design and results.
- Track record of managing cross-functional teams, for example in a tech company, consulting firm, or fast-paced startup.
- Experience engaging external partners such as frontier AI companies, government actors, or civil society organizations.
- Ability to work through ambiguity and deliver against complex, evolving goals on short timelines, building workflows where none exist.
- Ability to translate technical research for a range of audiences.
- Exceptional communication and relationship skills, low ego, openness to giving and receiving feedback.
Preferred qualifications (not required):
- Experience managing evaluations at scale or in a production context.