Member of Technical Staff - ML Platform Engineer

Preference Model • San Francisco, CA

1d

About The Position

Preference Model is a superintelligence data research company that builds reinforcement learning environments for training capable, better-aligned superintelligences. The company researches how to build better RL environments and has built environments for several frontier labs. It is backed by $16 million in seed funding led by a16z. Its founding team previously worked on Anthropic’s data team, building data infrastructure and datasets behind Claude.

As a Member of Technical Staff on the ML Platform team, you will own the systems that take an environment from a builder’s repository to a lab’s training run. This includes Karotte, the company’s open-source framework for defining RL environments; the internal platform for building, running, and reviewing environments on CPU, GPU, and TPU; and integrations that deliver environments to partner labs. You will ensure the Capabilities team’s work runs and ships cleanly, working on a small Platform team alongside environment builders and post-training researchers. The goal for the first six months is to make delivery to each partner lab mostly automated, so tasks marked ready are built, validated against lab requirements, and packaged without manual work. You may also take on other platform and infrastructure work as priorities change.

Requirements

• At least 5 years of experience building and owning ML platform engineering, ideally AI-related ML platform engineering.
• Strong software engineering fundamentals and a track record of building and operating production infrastructure or developer platforms used by other engineers.
• Hands-on experience with containers and orchestration such as Docker, Kubernetes, or similar.
• Experience with at least one major cloud platform, such as GCP or AWS.
• Working understanding of safely running untrusted code, including isolation, resource limits, and handling processes that fill disk space or refuse to exit.
• Experience owning a tedious process end to end, such as a migration, release process, or delivery pipeline, and automating as much of it as possible.
• Extensive hands-on use of coding agents such as Claude Code, Cursor, or Codex, and intuition for where models cut corners.
• Proficiency in Python, including writing libraries and APIs for other people to build on.

Nice To Haves

• Experience building RL environments, evaluation harnesses, or agent benchmarks such as Terminal-Bench or SWE-bench.
• Familiarity with RL post-training and inference stacks such as verl, Ray, SkyPilot, vLLM, or SGLang.
• Experience building sandbox or code-execution infrastructure, such as microVMs using Firecracker or gVisor, at an infrastructure startup or frontier lab.
• Experience as an early platform or infrastructure engineer at a startup, building the internal platform used by the rest of engineering.
• Experience as a forward-deployed or integrations engineer, getting a company’s software running in customers’ environments.
• Experience building developer tooling or internal platforms used by ML researchers, including outside a formal ML role.
• Experience maintaining an open-source developer framework with real external users.

Responsibilities

• Own integrations that deliver environments to partner labs, including building container images, matching lab submission formats, and automating pre-shipping requirement checks.
• Perform delivery work manually when needed and automate it where possible.
• Build and maintain Karotte, the open-source environment framework, including API design, releases, and migrations that keep existing environments working.
• Run and scale infrastructure for building, running, and grading environments across CPU, GPU, and TPU, including containerized rollouts, VM-based sandboxing, resource limits, and isolation.
• Build internal tools for environment builders, including a CLI, documentation, data catalog, access controls, and run monitoring.
• Work with environment builders and post-training researchers to turn recurring pain points into platform features and respond quickly to feedback.
• Take on other platform and infrastructure work as priorities shift.

Benefits

• Competitive cash and equity compensation (>90th percentile).
• Ownership and autonomy in a fast-moving startup environment.
• Opportunity to work alongside senior and staff engineers from frontier labs and infrastructure companies, plus top ML engineers.
• Health, vision, and dental benefits.
• 401(k) match.
• Lunch provided every day onsite.
• Weekly snack orders.
• Visa sponsorship and relocation support available.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed