Technical Program Manager, Research
Anthropic · San Francisco, CA | New York City, NY · San Francisco, CA
CHECKED 2H AGO
Frontier labonsite$365k–$435k
At a glance
Highlights
- Work across full model lifecycle from pre-training to safety and interpretability.
- Drive eval readiness for launches with honest, transparent reporting.
- Own RL environments and compute planning across major training runs.
What they offer
- Competitive compensation and benefits
- Optional equity donation matching
- Generous vacation and parental leave
- Flexible working hours
- Collaborative office space
Ideal candidate
High-agency, fast-learning program leader with ML research or engineering roots who thrives in ambiguity. You love connecting research, engineering and product to accelerate frontier AI and care deeply about safe, beneficial systems.
Requirements
- Bachelor's degree or equivalent combination of education, training and experience
- Background in ML research or engineering with several years building technical programs from scratch
- Fast learner able to ramp on unfamiliar technical domains quickly
- Resourceful, high-agency and able to navigate ambiguity and shifting priorities
- Track record of operational ownership including monitoring, incident response and optimization
- Can reason about technical tradeoffs and translate into clear decisions for leadership
- Excellent stakeholder management and ability to influence senior technical staff
- May require responding to incidents on short notice, including weekends
- remote policy
- Hybrid - in office at least 25% of the time
- visa sponsorship
- We sponsor visas and will make every reasonable effort if you receive an offer
Lab profile
About Anthropic
AI safety company building reliable, interpretable and steerable frontier AI systems.
What they build
- Claude
- Claude API
- Claude Code
Known for
- AI safety and alignment
- Model interpretability research
- Frontier LLM development
Key facts
- Founded in 2021
- Headquartered in San Francisco, California
- Backed by Amazon and Google