Forward Deploy Engineer

Design, build, and maintain sandboxed environments for AI agent training and evaluation Create containerized environments using Docker, VMs, and lightweight sandboxing solutions Develop tooling that allows agents to interact with terminals, browsers, APIs, and external tools Build observability, logging, monitoring, and data capture frameworks Partner with operations teams to create evaluation tasks and testing environments Manage CI/CD pipelines, automated testing, and environment reliability Prototype new environment types quickly based on evolving AI and client needs Work heavily across infrastructure, developer tooling, and agent evaluation systems

2+ years of professional Software Engineering experience Strong Python development experience Experience with at least one systems language (Go, Rust, or C++) Hands-on containerization/sandboxing experience (Docker, Podman, Firecracker, etc.) Experience building developer tools, CLI tools, or infrastructure automation, MDPs Reward shaping Episodes Observation/action spaces Experience working in Linux/terminal-heavy environments Strong debugging and troubleshooting skills Ability to read and implement from technical documentation, academic papers, and open-source repositories Built or maintained RL environments, Gymnasium/Gym PettingZoo OSWorld TerminalBench WebArena SWE-bench Tau-bench Browser automation experience (Playwright, Selenium, Puppeteer, etc.) AWS or GCP experience AI infrastructure, agent evaluation, or ML platform experience Open-source contributions Startup engineering background Experience at AI labs, AI tooling companies, or ML platform organizations