SF-Based Research Scientist, Agentic AI & Simulation

Patronus AI; United States; Remote; USD 175,000 - 300,000; Full time

Benefits offered by this job

Salary + equity

Vacation 15 days

Parental leave

Health, dental, and vision insurance

401(k) with matching

In-office meals

Wellness stipend

Equinox membership

Global offsites

Job summary

Patronus AI, based in San Francisco, seeks a Researcher to own foundational research at the intersection of reinforcement learning, simulations, and scalable oversight. You will shape how frontier models are trained, evaluated, and deployed, tackling open-ended questions with rigorous experiments and production-ready systems.

This highly autonomous role requires translating research into scalable infrastructure, writing production-level code, and contributing to the industry standard for robust

Qualifications

• Experience conducting independent research in reinforcement learning, NLP, agentic systems, evaluation, alignment, or related areas.
• Strong experimental skills, including experiment design, analysis, and interpretation of results.
• Clear written and verbal communication skills, including the ability to explain complex technical ideas succinctly.

Responsibilities

• Own ambitious research projects end-to-end, from identifying problems through design, execution, analysis, and production impact.
• Advance research in agent simulation, reinforcement learning, and scalable oversight.
• Design state-of-the-art simulation and RL environments for training and evaluating frontier agents.
• Train models and experiment with post-training algorithms, including GRPO and SFT; run ablations with open source models.
• Develop methods to understand and improve agent behavior across long-horizon tasks.
• Run rigorous experiments and turn findings into measurable outcomes.

Education

MS or PhD in Computer Science, Machine Learning, Statistics, Mathematics, or a related quantitative field

Tools

Python

Job description

Patronus AI, based in San Francisco, seeks a Researcher to own foundational research at the intersection of reinforcement learning, simulations, and scalable oversight. You will shape how frontier models are trained, evaluated, and deployed, tackling open-ended questions with rigorous experiments and production-ready systems.

This highly autonomous role requires translating research into scalable infrastructure, writing production-level code, and contributing to the industry standard for robust