About the company
Our client is a robotics company.
The role
Raydar is recruiting for this role on behalf of our client. Own machine learning work for physical robots, from collected data through to a learned control policy running on hardware. The role covers building data tooling, training and refining models, and making them perform reliably on real machines. The engineer also helps shape the group's expectations for handling datasets and running training jobs.
What you'll do
- Build data pipelines that ingest and validate large volumes of robot data each day.
- Fine-tune learned control models and maintain fast, dependable training workflows.
- Calibrate hardware and oversee data quality checks so models perform well on the local setup.
- Set the bar for how datasets are handled and how training runs are carried out.
- Track new robot learning research and adopt the approaches that prove useful on real systems.
Requirements
What we're looking for
- Zero to seven years of relevant experience in robotics or machine learning.
- Hands-on experience with physical robotic systems such as manipulators, drones or autonomous vehicles.
- Applied machine learning engineering background rather than purely academic research.
- Experience training and deploying learned policies on a robot that moves.
- Strong proficiency in Python.
- Working knowledge of PyTorch, JAX or a comparable deep learning framework.
- Depth in robot learning, including policy post-training and training pipelines.
- Openness to applying robotics skills to government-related projects.
Bonus points
- Experience at leading robotics or autonomous vehicle companies.
- Defense-tech experience.
- Background from a strong engineering university program.
- Proficiency in C++ or Rust.
- Low-level GPU programming experience.
- Experience with distributed training or evaluation infrastructure.
Benefits
Compensation and benefits
- Base salary: USD 150,000 to 250,000 per year
- Equity
- Medical, dental and vision coverage
Location and work model
- San Francisco, CA, United States
- On-site, 5 days in office per week, with Saturday work as needed
- Full-time