Frontier RL Research Intern: Design & Evaluate Environment

Traverse; San Francisco (CA); On-site; USD 30,000 - 42,000; Full time

Job description

Traverse is a research data lab building reinforcement learning environments for frontier AI labs. We focus on the non-deterministic, taste-dependent work that makes up most of the economy and that nobody else has figured out how to train models on. We work directly with the labs building the most capable models on earth as a thought partner. Backed by Y Combinator.

About the Role

As a Research Intern, you will work alongside the research team to design and evaluate RL environments for domains where expertise is hard to formalize. You'll get hands-on experience at the frontier of AI training, contributing to real projects that ship to partner labs.This is a high-impact internship for someone who learns fast, thinks carefully, and wants to do work that matters. No prior ML or AI experience is required — we care about intellectual curiosity and the ability to reason rigorously about hard problems.

In this role, you will

• Assist in designing and evaluating RL environments across new domains
• Run experiments to test hypotheses about reward modeling and environment quality
• Collaborate with domain experts to understand what mastery looks like in a given field
• Contribute to internal research documents and help refine Traverse's methodology
• Work directly with research scientists and engineers on active projects

Your background looks something like this

• Currently pursuing or recently completed a BS, MS, or PhD in any rigorous field
• Strong analytical and problem-solving skills
• Ability to learn quickly and work independently in an ambiguous environment
• Genuine curiosity about how expertise works across different domains

Bonus

• Familiarity with reinforcement learning, RLHF, or LLM evaluation