Senior Perception Engineer Machine Learning Operations - Agility Robotics

About The Role:

Our software team is growing rapidly and we are looking for talented engineers to join our small, agile, world-class team. In this role you will support efforts to build a scalable machine learning cluster and help deploy learned models on robot hardware that will enable Digit to autonomously interact with its environment and manipulate objects during real-world deployments with industry partners.

About The Work:

• Support training, optimization, and deployment of machine learning models on a large fleet of state-of-the-art humanoid robots
• Deploy a variety of machine learning models from cluster to robots to improve perception and task performance for specific customer environments and applications
• Review, order, and set up hardware for a scalable machine learning GPU cluster
• Set up cluster software and protocols for training and evaluating machine learning models
• Develop machine learning performance tests on secondary cluster that uses robot compute and GPU hardware
• Deploy machine learning models to robots in the field
• Work with other teams in software (Manipulation, Navigation, Planning, and Controls) to solve complex cross functional problems

About You:

• You have

• BS or MS in computer science or related field or have equivalent industry experience
• 5+ years of experience with industry standard ML data and model management tools
• Experience setting up compute systems for training ML models
• Experience deploying neural networks onboard robots with limited compute (e.g., TensorRT, ONNX, Pytorch Mobile, Tensorflow Lite, etc.)
• Familiarity with common machine learning libraries such as (but not limited to) PyTorch and Slurm
• Strong software experience in Python
• Strong experience with version control (Git)
• Ability to communicate technical information effectively both orally and in writing with other team members
• Ability to work independently and as part of a team with a demonstrated ability to establish and maintain effective working relationships
• Authorization to work in the USA

• You might also have

• Experience with robotics hardware and embedded systems
• Experience setting up compute systems for simultaneously training, validation, and deployment of hundreds to thousands of ML models
• Experience with Docker and Kubernetes
• Experience with GPU Programming (eg. CUDA, OpenGL)
• Experience with modern C++
• Exposure to standard agile and software development processes (issue tracking tools (Jira), feature branching, continuous integration and deployment