About the job
We're seeking research engineers to build infrastructure for breakthrough innovations in AI agents, reinforcement learning, and simulation environments. You will design and implement high-quality data pipelines, simulation systems, and tooling that enable cutting-edge agent research. You will work in an organization of world-class machine learning researchers and engineers. Our work powers technologies across the Apple ecosystem and is published in the most selective scientific journals and conferences.
Responsibilities
Passion for the mission: We're here to make something extraordinary. We seek whatever work is right and strive for the best possible results.
Modesty: The right answer is more significant than being right. We search for solutions as a team and value clear-eyed feedback.
Lean habits: You can't grow without limits. Time constraints and big goals encourage us to sharpen our focus and learn to make phenomenal decisions.
Qualifications
Minimum
5+ years of ML engineering experience building and maintaining data-intensive systems - including feature pipelines, training infrastructure, model serving, or evaluation frameworks.
Solid software engineering skills in complex systems - Fluency in Python. You deliver clean, well-tested code.
Hands-on experience with distributed ML systems - CI/CD at scale, distributed testing, or ML evaluation pipelines.
Strong quantitative and data skills - comfortable with SQL, statistical reasoning, and translating ambiguous signals into clear, actionable findings.
Proven track record shipping ML systems end-to-end - from problem framing and data curation through training, evaluation, deployment, and monitoring in production.
Preferred
Bachelors or Masters Degree in Computer Science, Engineering, Math, or Physics from a strong program.
2+ years at a company building AI products or agent systems.
Experience with job orchestration frameworks (Airflow, Prefect, Ray, etc.).
Familiarity with macOS/iOS development ecosystems.
Active personal interest in AI agents—you're already experimenting on your own time.