About the role
At Rhoda AI, we’re building the next generation of generalist intelligent robots. We own the full robotics stack from high-performance hardware and robot systems to the infrastructure and state-of-the-art foundation world models that control our robots. Our robots are designed to be generalists capable of operating in complex, real-world environments and handling long-tail edge cases, made possible by our cutting edge research and end-to-end system design. We've raised over $450M and are investing aggressively in model research, infrastructure, hardware development, and manufacturing scale-up to make generalist robotics a reality.
You'll own the safety requirements and safety evidence for the robot's perception path, end to end: from the sensors and their data links, through detection and decision logic, to the robot's response. You'll make sure the path is deterministic, auditable and explainable, and that it meets its safety targets.
What You’ll Do
- Algorithms: Own the end-to-end safety targets and performance of perception algorithms used in robot’s collision avoidance system, develop evidence that design meets them
- Requirements: Derive by engineering analysis system safety requirements for detection performance, classification accuracy, KPIs, identifying and effectively mitigating malfunctions, degradation in performance, and edge cases
- SOTIF: Develop detection and mitigation strategies within SOTIF framework and own SOTIF performance from developing requirements to analyzing VnV test data
- Calibration: Set safety requirements for calibration accuracy, drift detection and recalibration triggers, and analyze data for trends and drifts
- Error budgets: Set the safety limits for the sensor-to-actuator error budget, check that the allocation stays inside them, contribute to trade studies and influence hardware/software designs
- Software safety: Create safety concept and the safety requirements for its implementation, working closely with Software Engineering team, achieving safety-compliant software implementation
- Sensor hardware: Define safety requirements for safety-related sensors (diagnostics, fault detection, failure rates), review supplier safety manuals and be a critical partner in sensor selection, and specify the qualification evidence needed and drive verification.
- Safety work products: Generate work products towards Robot Safety Case, drive reviews and approvals with an NRTL
What We’re Looking For
- Algorithms: Expertise in deterministic, auditable, explainable computer vision algorithms used for perception
- Requirements: Experience in analytically and empirically deriving engineering requirements with numerical pass/fail constraints to physical performance expressed in distance, time, speed, etc.
- SOTIF: Proven record of developing SOTIF work products for a safety-related robotic or automotive system
- Calibration: Experience creating calibration fixtures, defining calibration processes and driving failure and trend analysis of field data
- Error budgets: Experience deriving and allocating error budgets, experience in developing analysis backed by data to inform trade studies to achieve compliance with error budgets
- Software safety: Strong background in safety-critical embedded software, real-time operating systems, computer vision algorithms, and safety-compliant C/C++ development
- Sensor hardware: Experience with Lidars, Cameras, Time-of-Flight sensors, Radars, Ultrasonics, other sensing technologies and related sensor pipeline hardware and state of art methodologies used in safety-related context
Preferred
- MS or PhD in Computer Science, Electrical Engineering or Robotics with focus on Computer Vision and Embedded Systems
- Experience creating work products for Functional Safety and SOTIF and successfully passing an audit by an NRTL
- Experience with safety-rated compute hardware with strong vision processing capabilities
- Experience designing safety architectures and documenting requirements within the context of IEC 61508 or ISO 26262
- Experience with embedded software system developed with compliance to MISRA C, ISO 26262-6 or IEC 61508-3
- Perception Sensors Functional Safety background from automotive industry