Software Engineer - Reinforcement Learning Infrastructure, MSL (Technical Leadership)

Meta is seeking a hands-on technical leader to advance Personal Superintelligence within Meta Superintelligence Labs (MSL). This role focuses on building and maintaining RL infrastructure that can support the latest evals, RL datasets, and experimentation at the frontier of consumer and enterprise agents, coding, and beyond. In this role, you will work across the research stack as well as connect the research stack to product by building black-box RL infrastructure. You will collaborate across research, product engineering, data, and infrastructure to identify high-leverage opportunities and deliver tangible impact.

Responsibilities

Maintain RL black-box environment verticals for Muse, MetaAI, and other internal products Ensure that we can securely run the most challenging agentive evals and RL problems Think holistically about security, development environments for RL, their interactions with various VM solutions, and our core RL infrastructure framework

Qualifications

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience Bachelor’s or Master’s degree in a relevant technical field, or equivalent practical experience 6+ years of industry experience in machine learning engineering, AI research, software engineering, or a related field 4+ years of providing technical leadership for complex, multi-person projects Experience working on reinforcement learning infrastructure at scale - PPO/GRPO-family algorithms, RLHF/RLAIF, or agentic RL with long-horizon, tool-using rollouts Strong Python, plus proficiency in a systems language (C++, Rust, or Go) Track record of solving ambiguous, real-world problems and delivering measurable impact within defined timelines Ability to work independently, lead across functions, and adapt quickly as evidence and priorities evolve Experience building sandboxed execution or environment infrastructure Experience building platforms used by research teams, with a track record of good API/abstraction design Contributions to open-source machine learning infrastructure Comfort with ambiguity and rapidly shifting research priorities