Member of Technical Staff
Core
Design, build, and maintain sandboxed, reproducible environments for training and evaluating agentic AI systems, including terminal emulators, browser automation, and tool-augmented workspaces.
Role type
Senior IC infrastructure engineer (RL environments)
Builds
Sandboxed execution environments, containerized systems, and observability layers for AI agent training loops
Domain
Artificial Intelligence / Reinforcement Learning / Infrastructure
Deliverable
production ML models
Required skills
Python, Go/Rust/C++, containerization (Docker, Podman, Firecracker), RL concepts (MDPs, reward shaping), CLI/API development, debugging across process boundaries, reading academic papers
Preferred skills
RL environment frameworks (Gymnasium, PettingZoo), agentic evaluation frameworks (SWE-bench, WebArena, OSWorld), cloud infrastructure (GCP, AWS), open-source contributions
Technologies
Python, Go, Rust, C++, Docker, Podman, Firecracker, TerminalBench, OSWorld, Tau-bench, GCP, AWS
Responsibilities
Design and build sandboxed RL environments for agentic AI training; Develop reproducible, containerized execution environments; Integrate with open-source agentic tooling and custom CLI/API harnesses; Build instrumentation and observability layers for training runs; Collaborate on task curricula and evaluation protocols; Own environment deployment, CI/CD pipelines, and reliability monitoring
Seniority
Mid-level, hands-on IC
