Research Engineer, Large Behavior Models - Computer Vision
tri
Los Altos • fulltime
Posted on: 6/5/2025
Required Skills:
PythonRoboticsMachine Learning
Job Description:
Research Engineer - AI-Powered Robots
We’re looking for a motivated research engineer to help build the next generation of AI-powered robots through high-fidelity simulation and large behavior models. You’ll work at the intersection of machine learning, computer vision, robotics, and physics-based simulation to enable scalable, generalizable robot policy development. Experience with simulation or embodied systems (robots, autonomous vehicles, etc.) is a strong plus. If our mission of revolutionizing robot learning through simulation and large behavior models resonates with you, we’d love to talk about how we can build it together.
Responsibilities
- Develop and manage physics-based robot simulation environments using Drake, enabling scalable training and evaluation of learning-based behavior models in realistic, physically grounded scenarios.
- Integrate and validate learned policies in simulation, assessing real-world applicability, generalization, and performance across diverse environments, tasks, geometries, and sensor viewpoints.
- Build, improve, and robustify end-to-end integrated ML pipelines for training multimodal (language, images, 3D, video, actions) models at scale.
- Train, fine-tune, and serve robot foundation models with a strong MLOps approach.
- Build processes for integrating collaboration-produced and open-source advancements and code into our internal stack.
- Collaborate with internal research scientists and our partner labs at top academic institutions and Toyota research labs to drive groundbreaking research at scale.
Qualifications
- 2+ years of professional engineering experience at an AI/ML-focused organization.
- Strong proficiency in Python and experience with simulation frameworks such as Drake, PyBullet, MuJoCo, or similar.
- Hands-on experience with robotics simulation, reinforcement learning, or large-scale machine learning.
- Familiarity with the latest methods in behavior learning and/or computer vision.
- Experience integrating ML models into simulated or real-world environments.
- Extensive practical experience with PyTorch.
- Ability to alternate between rapid prototyping and production-quality implementation.
- Demonstrated understanding of software engineering best practices, including testing, CI/CD, and documentation.
Bonus Qualifications
- Experience deploying models on embodied systems/robots.
- Experience working in mixed teams of research scientists and engineers.
- Exposure to MLOps tools and infrastructure (e.g., Docker, EC2, S3, Sagemaker).
- Experience with Bazel.