Perception Engineer, Robot Safety

Job description

Job responsibilities

• Build the perception systems that keep people safe: Detection, tracking, and monitoring from lidar and cameras, running on robots in production
• Own safety-critical code: The software path from a sensor reading to a stopped robot, including its tests, its failure modes, and its behavior when something upstream fails
• Design and review safety-relevant systems: Set the design for how a robot decides it is safe to move, and review changes across the codebase that affect that decision
• Test for weaknesses: Use replayed field data, constructed scenarios, and hardware faults to find where the system fails, and fix what you find
• Work across product teams: Join the team building something that affects safety and build it with them
• Help set the strategy: Contribute to how Tutor keeps its robots safe and deployable as the fleet and the applications grow, including the evidence we give customers

Job requirements

• Experience shipping perception or monitoring software on robots or vehicles: Lidar or camera pipelines, detection, tracking, or sensor fusion, running on hardware in production
• You have owned code where a bug could injure someone: You know how that changes design, testing, and review
• Strong software engineering: Python and C++ or similar, real-time constraints, and code that other people can verify
• Some controls knowledge: Stopping distances, latency budgets, and how a perception decision becomes a motion command
• You fix what you find: When you find a weakness, you propose the fix and usually build it