AI Tooling Engineer - Model Performance & Evals

xAI is seeking an AI Tooling Engineer focused on model performance and evaluations. The role involves building tools to benchmark, evaluate, and improve AI models' reasoning and performance, designing pipelines for agent testing and iteration at scale, collaborating with researchers to quickly turn insights into deployable systems, leveraging GPU resources to prototype and scale advanced solutions, and innovating towards AI that deeply understands the universe. Candidates should have experience shipping ML-adjacent tools or workflows, expertise in evals, data pipelines, or model optimization, and full-stack or systems skills bridging research and production. The tech stack includes Typescript, Python, Rust, React, Express, and PostgreSQL. The position is based in the Bay Area (San Francisco and Palo Alto), with candidates expected to be local or open to relocation. The interview process includes a phone interview followed by four technical interviews covering coding, front-end, systems, and a presentation of past work. All interviews are conducted via Google Meet. xAI values engineering excellence, strong communication, initiative, and a flat organizational structure.