Perception Engineer, Robot Safety

Tutor Intelligence•Watertown, MA

11d•$150,000 - $300,000•Onsite

About The Position

We are looking for an engineer to build and own the perception systems that keep people safe around our robots. These systems include sensors that detect a person and software that determines the robot's response. This is an engineering role where you will design and build perception and monitoring systems, own safety-critical code end-to-end, and be responsible for the design and review of systems that determine robot movement. You will collaborate across product teams, joining whichever one is building something that affects safety, and will proactively test systems for weaknesses. You will also contribute to shaping Tutor's overall safety strategy, including understanding risk assessments and customer standards. The core of this role is engineering, with training provided for the formal aspects of safety.

Requirements

Experience shipping perception or monitoring software on robots or vehicles, including lidar or camera pipelines, detection, tracking, or sensor fusion, running on production hardware.
Experience owning code where a bug could cause injury, demonstrating an understanding of how this impacts design, testing, and review.
Strong software engineering skills in Python and C++ or similar, with experience in real-time constraints and verifiable code.
Some controls knowledge, including understanding stopping distances, latency budgets, and how perception decisions translate to motion commands.
A proactive approach to fixing identified weaknesses.

Nice To Haves

Exposure to safety standards for robots or machinery, such as ISO 10218, ISO 13849, or ANSI/RIA R15.06.
Experience with safety-rated sensors or safety validation of a detector.
Experience in warehouses, manufacturing, or other environments where robots and people share space.
Experience with simulation and replay tooling for testing perception at scale.

Responsibilities

Build the perception systems that keep people safe, including detection, tracking, and monitoring from lidar and cameras, running on production robots.
Own safety-critical code, encompassing the software path from sensor reading to robot stoppage, including tests, failure modes, and behavior during upstream failures.
Design and review safety-relevant systems, setting the design for how a robot determines it is safe to move and reviewing codebase changes affecting this decision.
Test for weaknesses using replayed field data, constructed scenarios, and hardware faults, and fix identified issues.
Work across product teams, joining teams building safety-affecting components and contributing to their development.
Help set the strategy for how Tutor ensures robot safety and deployability as the fleet and applications grow, including providing evidence to customers.