Stretch is Boston Dynamics’ mobile manipulation robot built to handle multiple applications in warehouse environments.
We are seeking a technically talented software engineer with a background in AI/ML to help advance Stretch’s manipulation capabilities through the application of data-driven techniques.
As a member of the Stretch Manipulation team, you will be a key contributor to tackling some of the most challenging technical problems in robotics. Your role will focus on the hands-on implementation of data processing and model evaluation pipelines, and model development and deployment in real-world settings. Success in this position requires a highly collaborative mindset, ambitious ideas, and rapid and meticulous execution within a dynamic, fast-paced team.
To thrive in this role:
You are keen on applying advancements in ML/AI (e.g., reinforcement learning, imitation learning) to robotics applications: you stay in the know on latest innovations in the ML/AI space, actively look for ways to apply them, and develop and deploy learned models to enhance Stretch’s manipulation capabilities and increase robot performance.
You recognize the value and necessity of ensuring data pipelines, evaluation frameworks, etc. scale as our business and deployed robot fleet grows: you’re excited to collaborate closely with infrastructure engineers to build and iterate on AI/ML development systems, spanning simulation, data collection, model training, validation, and deployment on real robots.
You’re excited to collaborate with members of the manipulation team (and adjacent teams) to integrate learned components into the broader autonomy stack.
You’re eager to contribute to cross-company efforts enabling scalable AI experimentation and deployment.
You have a strong desire and ability to mentor and up-skill other engineers (for more senior applicants).
We’re Looking For
Bachelor's, Master's, or Ph.D. in Computer Science, Machine Learning, or a related field; additional industry experience is a plus.
Proficiency with modern ML frameworks/ecosystem (e.g., PyTorch, Jax, TensorFlow, RLlib, ONNX, inference servers like NVIDIA Triton).
Familiarity with modern model architectures and learning techniques and their applications/tradeoffs.
Strong programming skills in Python and C++ within a Linux environment; familiarity with modern agentic programming workflows (e.g., claude code, copilot, etc.).
Excitement for integrating, running, and evaluating their learned policies on-robot.
Excellent communication and collaboration skills.
Preferred
Experience shipping learned models in a production environment.
Experience in robotics or warehouse automation.
Experience building and leverage high-fidelity simulation environments (e.g., Isaac Sim, MuJoCo) or augmented/virtual reality (AR/VR).
Experience training and deploying RL policies for complex behaviors on real robots or simulated characters; experience with diffusion policy, ACT, or other modern BC architectures.
Publications in top-tier ML conferences (e.g., NeurIPS, ICML, ICLR, CoRL).
The base pay range for this position is between $192,000 to $255,000 annually. Base pay will depend on multiple individualized factors including, but not limited to internal equity, job related knowledge, skills and experience. This range represents a good faith estimate of compensation at the time of posting. Boston Dynamics offers a generous Benefits package including medical, dental vision, 401(k), paid time off and a annual bonus structure. Additional details regarding these benefit plans will be provided if an employee receives an offer for employment.