Human Computer Lab
Perception Engineer
About this role
Human Computer Lab seeks a Perception Engineer to develop computer vision and machine learning systems that enable their expressive robot to understand people and environments in real time. You'll own the full pipeline from data collection through model deployment, building perception capabilities that work reliably in real-world conditions.
What you'll do
- Design, train, and deploy computer vision models for person detection, tracking, pose estimation, gesture and emotion recognition
- Develop data collection, labeling, and evaluation pipelines to improve model performance and reliability
- Optimize models for real-time inference on edge hardware while balancing latency, accuracy, and power constraints
- Integrate multi-modal perception systems using RGB cameras, depth sensors, IMUs, and audio inputs
- Debug perception failures systematically and drive data-driven improvements to close feedback loops
- Support integration between perception, behavior, controls, and firmware across the robot platform
What they're looking for
- Python and C++
- Machine learning frameworks (PyTorch, TensorFlow, ONNX)
- Computer vision (object detection, tracking, pose estimation, segmentation)
- Edge device optimization and real-time inference
- Dataset management, labeling, and curation
- Multi-modal sensor integration
- ML systems debugging and problem diagnosis
- Robotics fundamentals
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Human Computer Lab
Human Computer Lab is a robotics startup building LeLamp, an expressive consumer robot designed to be responsive and lifelike. The company is hiring ML engineers, controls engineers, and electrical engineers to develop the intelligence systems, motion control, and hardware that bring the robot to life from research through production deployment.
View all jobs at Human Computer LabLikely interview questions
- Walk us through a computer vision project where you deployed a model to an edge device—what were the key optimization tradeoffs you made?
- Describe your experience debugging a ML system failure. How did you distinguish between data, model, and deployment issues?