Software Engineer - Atlas Behavior Learning

We are seeking talented software engineers to join our Atlas behavior learning team.  You will develop robot system software, applications, machine learning infra, and tools as part of our R&D effort teaching Atlas the skills it needs to act in the real world.  We are tackling the most challenging problems in humanoid robot manipulation and locomotion through state-of-the-art methods in learning and learning from demonstration.  In this project you will develop teleop interfaces, robot command interfaces, simulation and mocap tools, etc. to collect human demonstration datasets and train large behavior models for Atlas.  You will work on these hard challenges in close collaboration with roboticists, researcher scientists, DevOps engineers, and data annotation teams.  Your contributions will help us build new robot technologies for Atlas and other R&D efforts at Boston Dynamics.

How you will make an impact:

• Actively participate in ongoing research initiatives, collaborating closely with Research Scientists and Infrastructure Engineers

• Write software for data logging, data pipelines, training pipelines

• Work with robot simulation software to develop scenarios, tests, and data collection tools

• Develop visualization and analysis tools

• Work with VR and mocap devices and write code for teleoperation interfaces

• Work with annotation teams and develop data annotation tools

• Analyze and track failures in experimental robot runs.

• Maintain and improve the code repository, ensuring its ongoing maintenance and improvement to meet evolving needs (CI/CD, testing, deployment)

• Develop tools to query and gain insights into data, training, and testing performance metrics

Desirable experience:

• 3+ years of experience developing software for robotics or related fields

• Advanced knowledge in Python programming, familiarity with C++

• Familiarity with popular ML frameworks

• Experience with robotics or related problems: 3D transformations, kinematics

• Experience with ML robotics datasets, robot middleware systems, pubsub, and common sensor modalities

• Experience with cloud infrastructure, storage, databases

• Experience with docker and job orchestration for cloud and on-premise GPU clusters

We are interested in every qualified candidate who is eligible to work in the United States. However, we are not able to sponsor visas for this position.

#LI-JM1