Strong Python and C++ programming skills, including debugging and performance profiling.
Experience training and evaluating computer-vision models with frameworks like PyTorch.
Experience developing on embedded Linux and deploying models on resource-constrained hardware.
Understanding of quantization, model conversion, and inference optimization on CPUs, GPUs, or NPUs.
Ability to evaluate accuracy, false positives, and missed detections across varying conditions.
For the lead position: experience making architecture decisions and guiding engineers.
Responsibilities
Train and evaluate computer-vision models for deployment on embedded devices.
Build efficient C++ inference pipelines and integrate them with embedded Linux software.
Optimize models and runtime performance for latency, memory, and power consumption.
Test performance on representative data and hardware, investigating errors across preprocessing, inference, and integration.
For the lead position: set architecture and evaluation standards while continuing to develop models and software.
Benefits
Encouragement to apply even if not meeting 100% of qualifications, promoting diversity and inclusion.
Opportunity for technical leadership in model development and deployment.
Full Job Description
The Role You'll put advanced computer vision into the hardware that captures experience for robot learning. You'll train models, make them run on embedded devices, and carry their accuracy from controlled evaluation into real operating conditions, all within strict limits on latency, memory, and power. The role can include technical leadership of model development and deployment, and either way you'll be directly responsible for implementation and evaluation.
What You'll Do
Train and evaluate computer-vision models for deployment on embedded devices.
Build efficient C++ inference pipelines and integrate them with embedded Linux software.
Optimize models and runtime performance for latency, memory, power consumption, and sustained operation.
Test performance on representative data and hardware, and investigate errors across preprocessing, inference, and integration.
For the lead position: set architecture and evaluation standards while continuing to develop models and software.
What You Bring
Must-have:
Strong Python and C++ programming skills, including debugging and performance profiling.
Experience training and evaluating computer-vision models with a framework such as PyTorch.
Experience developing on embedded Linux and deploying models on resource-constrained hardware.
Understanding of quantization, model conversion, and inference optimization on CPUs, GPUs, or NPUs.
Ability to evaluate accuracy, false positives, and missed detections across changing lighting, motion, and operating conditions.
For the lead position: experience making architecture decisions and guiding engineers through model development and deployment.
Nice-to-have:
Experience with accelerator runtimes, custom operators, or camera and video pipelines.
Experience with dataset curation and rigorous evaluation of computer-vision models.
A Note on Applying
Studies show women and candidates from underrepresented groups often only apply when they meet 100% of the listed qualifications, while others apply after meeting 60%. If you don't check every box above but believe you can do the job, we encourage you to apply - we're looking for capability and trajectory, not a perfect checklist match.
