Staff Machine Learning Engineer, VLA Training (US)

We are looking for a Staff Machine Learning Engineer with 6+ years of experience to own the training, experimentation, and deployment of vision-language-action models that power autonomous robotic systems for U.S. Defense manufacturing. You'll join a fast-moving, early-stage team that has already closed seven figures of revenue with a tier-one defense prime and is scaling rapidly. This is a hands-on, applied role, not pure research, where you'll work to increase model reliability and speed for real-world robotic deployments. You'll build new systems from scratch, with the ability to test models on physical robots almost immediately. If you want to ship production-ready ML that moves real hardware in real environments, this is the opportunity.

What will you be doing?

• Train and iterate on vision-language-action models to hit reliability and speed targets for an active customer deployment

• Experiment with training data, recipes, and parameters to identify the levers that improve model success rates

• Build and scale distributed training infrastructure across multi-GPU clusters to support rapid experimentation

• Own the end-to-end model lifecycle, from dataset construction and training to evaluation and on-robot validation

• Scale experimentation capacity across multiple workflows as the company expands its operations