Perception Engineer
Human Computer Lab2 months ago
Toronto, Canada or San Francisco, CA, USAMid Level
Responsibilities
- Design, train, and deploy perception and computer vision models for detection, tracking, pose estimation, gesture recognition, emotion recognition, and scene understanding.
- Develop data collection, labeling, evaluation, dataset management, debugging, and monitoring tools and pipelines.
- Build robust perception systems for varying lighting conditions, environments, and user behaviors.
- Optimize models for real-time inference on edge hardware while balancing latency, accuracy, and power constraints.
- Combine RGB cameras, depth sensors, IMUs, audio inputs, and other modalities into multimodal perception systems.
- Integrate perception with behavior, controls, and firmware systems.
- Research state-of-the-art architectures and techniques for improving robotic awareness and interaction.
- Debug perception failures systematically and improve performance through data-driven iteration.
- Define and evolve the platform’s perception architecture as the robot gains capabilities.
Requirements
- A degree in Computer Engineering, Computer Science, Robotics, Electrical Engineering, or a related field is required.
- At least 3 years of hands-on experience is required.
- Proficiency in Python and comfort working in C++ when required.
- Experience training and deploying machine learning models with PyTorch, TensorFlow, ONNX, or similar tools.
- Experience with computer vision problems including object detection, tracking, pose estimation, segmentation, gesture recognition, or activity understanding.
- Experience deploying models to edge devices and optimizing performance for real-world systems.
- Understanding of collecting, labeling, and maintaining high-quality datasets.
- Ability to work in evolving architectural environments, document clearly, and drive issues to resolution.
- Strong debugging intuition and ability to distinguish data, model, and deployment problems.
- Ability to translate research into practical systems and evaluate accuracy, latency, robustness, power, and cost tradeoffs.
- Ability to take end-to-end ownership from data collection and training through deployment and monitoring.
- Ability to collaborate across software, hardware, controls, and AI in a small team.
Categories
ML EngineeringRobotics
About Human Computer Lab
Human Computer Lab is an independent research lab focused on human–robot interaction. The team designs and builds expressive robots and interaction systems that aim to make technology more legible and emotionally intuitive, combining hardware prototyping, software, and behavioral design. It operates as a privately held studio, working on in‑house experiments and collaborations that translate research into interactive products.