The company builds autonomous robotic systems for defense customers and is hiring a Staff Machine Learning Engineer to lead work on vision-language-action models.
The role is hands-on and centers on making models more reliable and faster for real-world deployment, with chances to test models on physical robots.
Responsibilities
Take ownership of the full model lifecycle, from building datasets and training through evaluation and validation on real robots.
Train and refine vision-language-action models until they meet the reliability and speed targets set for customer deployments.
Test different training data, recipes, and parameters to raise model success rates.
Design and scale distributed training infrastructure running on multi-GPU clusters.
Expand experiment capacity across several workflows as operations grow.
Requirements
You have worked six or more years as a machine learning or robotics engineer.
You hold a bachelor's degree in computer science, machine learning, robotics, electrical engineering, or a related field, or can show equivalent practical experience.
You have trained deep learning models in PyTorch or JAX and set up and run distributed training across multi-GPU clusters.
You have worked with transformer-based, multimodal, or generative models and shipped machine learning systems that run on physical hardware.
You can debug across the full stack and have independently delivered working systems from loosely defined projects.
Experience with VLA, VLM, or robotics foundation models is beneficial, and knowledge of imitation learning, diffusion policies, or behavior cloning is also beneficial.
You are familiar with ROS or ROS 2, MuJoCo, Isaac Sim, or Gazebo.
U.S. citizenship is required.
Conditions
This is a full-time position, not a contractor role.
Base salary runs from $250,000 to $300,000 per year.