About the role: Are you passionate about making robotics and AI workloads run reliably at the edge? Do you enjoy hands-on engineering across robots, cameras, GPUs, containers, embedded systems, and real-time operational constraints?
We are seeking a Robotics Engineer - Edge Runtime to help build and optimize the edge software layer for the Physical AI Runtime. This engineer will work on edge development, robotics integration, GPU acceleration, model optimization, NVIDIA runtime technologies, Jetson Thor and related acceleration platforms, and the practical performance work needed to run AI-assisted robot workloads near machines.
What you will do: - Bring up edge runtime software on supported robotics, GPU, and industrial edge compute targets.
- Integrate robot, camera, sensor, inference, and runtime services into deployable edge configurations.
- Profile and optimize latency, throughput, GPU utilization, memory movement, startup time, and resource contention.
- Support NVIDIA runtime technologies, Jetson-class platforms, GPU acceleration, and model optimization paths.
- Build automation for provisioning, profiling, diagnostics, health checks, and repeatable performance testing.
- Debug end-to-end failures across hardware, drivers, containers, networks, inference services, and robot interfaces.
- Work with ML Systems, SIMATIC AX, Robotics Integration, QA, and Perception engineers to keep edge behavior predictable.
What you will bring: - Bachelor's or Master's degree in Robotics, Computer Engineering, Electrical Engineering, Software Engineering, Mechatronics, Computer Science, or a related field.
- 3+ years Strong experience building robotics, edge, embedded, or performance-sensitive runtime software.
- Strong programming skills in C++, Python, or similar languages used for robotics systems, runtime services, and hardware integration.
- Experience with Linux, containers, device drivers, networking, IPC, system services, profiling tools, and deployment automation.
- Practical understanding of GPU acceleration, memory movement, latency profiling, throughput tuning, and resource budgeting.
- Experience integrating robots, cameras, sensors, edge compute, and runtime services into a working system.
- Ability to debug failures across hardware, drivers, network paths, containers, inference services, robot interfaces, and runtime orchestration.
- Experience defining health checks, resource budgets, startup behavior, degraded modes, and recovery behavior for edge workloads.
- Comfortable working hands-on with lab equipment, prototype cells, imperfect hardware, and evolving integration requirements.
What sets you apart: - Experience with NVIDIA Jetson, Jetson Thor, CUDA, TensorRT, DeepStream, Triton, Isaac, or comparable acceleration and robotics tooling.
- Experience deploying VLM, VLA, perception, or robot-learning workloads on edge GPUs or constrained compute targets.
- Knowledge of Industrial Edge, embedded Linux, real-time systems, fieldbus or industrial networks, robot controllers, and camera interfaces.
- Experience optimizing model execution through quantization, batching, graph optimization, memory layout, runtime selection, or hardware-specific tuning.
- Familiarity with UR robots, robot APIs, ROS or ROS2, proprietary robot interfaces, trajectory execution, and safety-rated boundaries.
- Ability to distinguish benchmark performance from end-to-end cell performance.
- Strong practical judgment about what must run on edge, what can run elsewhere, and how to keep the runtime supportable.
Work environment: - This role includes hands-on work with robot cells, edge computers, GPUs, cameras, sensors, and prototype hardware.
- Hybrid work is supported, with on-site presence expected for hardware bring-up, profiling, integration debugging and reference-cell validation.
Salary is commensurate with experience, and ranges between $ 90,000 CAD - $ 140,000 CAD, excluding bonus and benefits. In addition to base salary, this role includes eligibility for an annual discretionary bonus of 10% of base salary, based on Company performance metrics.
