At Boston Dynamics, we are developing the next generation of intelligent robots capable of operating in complex, human-centric environments. A critical component of this effort is creating a rich, accurate, and real-time understanding of the world around the robot, which serves as the foundation for our Reinforcement Learning (RL) policies. We are seeking a talented software engineer to join our team and help us build the essential bridge between the robot's perception of the world and its ability to act within it.
As a RL Awareness Research Scientist on the Atlas Controls team, you will be responsible for integrating the robot's 3D world model with our advanced RL control stack. Your work will be crucial in enabling our robots to perform complex manipulation tasks by providing the necessary environmental context for our learning algorithms. This is a unique opportunity to work at the intersection of robotics, 3D geometry, and machine learning, with a direct impact on the capabilities of our robots.
What You'll Do:
Design, build, and maintain the software interfaces that connect our robot's real-time 3D world model to our RL-based control systems.
Develop and optimize the data pipelines that provide our RL policies with the necessary environmental information to make intelligent decisions.
Collaborate with our perception and controls teams to ensure that the environmental model is accurate, efficient, and tailored to the needs of our learning algorithms.
Implement and test new features that enable our robots to better understand and interact with their surroundings.
We're Looking For:
Ph.D. in a robotics-related field involving reinforcement learning, or a Master's degree with 3+ years of relevant professional experience.
Direct experience with Reinforcement Learning and a solid understanding of other machine learning techniques.
Strong proficiency in C++ and Python.
Experience with 3D geometry, transformations, and data structures.
A solid understanding of robotics fundamentals, including kinematics and coordinate frames.
Nice-to-have:
Familiarity with simulation environments (e.g., Isaac Sim, MuJoCo).
Experience working with large-scale 3D data from sensors such as cameras and LiDAR.
A passion for building robust and reliable software for real-world robotic systems.
Familiarity with modern software development practices, including version control, testing, and CI/CD.
#LI-JM1
Similar Jobs
What We Do
Boston Dynamics builds advanced mobile manipulation robots with remarkable mobility, dexterity perception and agility. We use sensor-based controls and computation to unlock the potential of complex mechanisms. Our world-class development teams develop prototypes for wild new concepts, do build-test-build engineering and field testing and transform successful designs into robot products. Our goal is to change your idea of what robots can do.
