Member of Technical Staff

Traverse; San Francisco (CA); On-site; USD 190,000 - 260,000; Full time

Job summary

Traverse is a research data lab building reinforcement learning environments for frontier AI labs. We focus on non-deterministic, taste-dependent work and train models for real-world tasks.

You will design reward functions, evaluation pipelines, and the tooling needed to scale environments across domains. As a Member of Technical Staff, you will own problems end-to-end, work across the full stack, and help shape the company's technical direction as an early team member.

Qualifications

• Exceptional software engineering fundamentals and ability to ship production-quality code quickly.
• Comfort with ambiguity and a bias toward building rather than planning.
• Track record of owning complex technical projects end-to-end.
• You learn fast and go deep on whatever you're working on.

Responsibilities

• Design and build RL training environments across multiple verticals.
• Build infrastructure and tooling that enables environment quality to scale across domains.
• Work directly with frontier AI labs to integrate environments into their training pipelines.
• Own problems end-to-end, from scoping through implementation through evaluation.
• Shape the technical direction of the company as an early team member.

Skills

Software engineering basics

Ambiguity tolerance

End-to-end ownership

Fast learner

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 Member of Technical Staff, you will build the core infrastructure and environments that train AI models to do real work. You'll operate across the full stack, from designing reward functions and evaluation pipelines to building the tooling that lets us scale environment quality across new domains without it falling apart.This is not a narrowly scoped role. You'll work on whatever matters most, and what matters most changes as we move into new domains and take on new lab partnerships. We care about raw engineering ability and taste, not credentials or keywords on a resume. No prior ML or AI experience is required.

In this role, you will

• Design and build RL training environments across multiple verticals, working closely with domain experts and ML engineers
• Build infrastructure and tooling that enables environment quality to scale across domains
• Work directly with frontier AI labs to understand their training pipelines and integrate our environments into their workflows
• Own problems end-to-end, from scoping through implementation through evaluation
• Shape the technical direction of the company as an early team member

Your background looks something like this

• Exceptional software engineering fundamentals and ability to ship production-quality code quickly
• Comfort with ambiguity and a bias toward building rather than planning
• Track record of owning complex technical projects end-to-end
• You learn fast and go deep on whatever you're working on

Bonus

• Familiarity with ML training pipelines, RL concepts, or LLM post-training
• Experience building evaluation frameworks, reward models, or RL environments