Staff Autonomy Engineer

In this role, you will:

Own the end-to-end architecture and hands-on development of Chef's autonomy stack — perception, action prediction, closed-loop control, and sensor fusion — from prototype through production deployment
Design and implement learning-based systems that predict and execute robot actions from visual and operational observations, generalizing across diverse ingredient types with minimal per-item tuning
Build closed-loop control systems that integrate vision, weight feedback, force-torque sensing, and trajectory planning for robust real-world manipulation
Develop rapid onboarding capabilities — algorithms and tooling that enable new meal types and ingredient configurations to be deployed quickly from minimal demonstrations
Instrument autonomy systems for production monitoring, failure detection, and continuous improvement; build data collection pipelines that feed back into model retraining
Define objective qualification criteria for autonomy subsystems; drive reliability and performance improvements using production data across the deployed fleet
Investigate and integrate the latest advances in robot learning, foundation models, and manipulation research — continuously evaluating what's ready to move from research into production
Mentor senior engineers, lead design reviews, and help build a team capable of scaling Chef's autonomy stack to hundreds of deployed systems

What You Bring:

Demonstrated experience shipping a complete autonomy or manipulation system to real-world deployment — not just simulation or research prototypes
Deep technical expertise across at least two of: perception systems, visuomotor control, action prediction, sensor fusion, or motion planning
Strong software engineering fundamentals in Python and C++; experience building production-quality, maintainable systems across research and deployment codebases
Track record of technical leadership: driving architecture decisions, leading cross-functional projects, and raising the engineering bar on a team
Comfort with ambiguity and urgency — able to set direction and execute in a fast-moving startup environment

Nice-to-have:

Experience with food, agricultural, or consumer goods robotics where object variability is high
Experience with ML frameworks (PyTorch preferred) and the full model lifecycle — training, evaluation, deployment, and field iteration
Familiarity with foundation model approaches for manipulation: diffusion policies, VLAs, world models, JEPA-style architectures
Background in sim-to-real transfer, domain randomization, or synthetic data generation for manipulation
Experience with ROS and real-time robotics middleware
Prior startup technical leadership experience

Chef Robotics is solving one of the hardest problems in AI: bringing intelligence into the physical world.

Our robots are already operating in production facilities every day, generating the real-world data that powers the next generation of embodied AI. If you want to build technology that leaves the lab, ships to customers, and transforms an industry, Chef is the place to do it.