Motion Control and Planning Intern at Apptronik in Austin, TX

Summary: Apptronik is a human-centered robotics company developing AI-powered robots to support humanity in every facet of life. As a Motion Control and Planning Intern, you will work directly with a senior mentor to plan and execute a high-impact project that showcases new functionality on physical robot systems, bridging the gap between simulation and reality. Responsibilities: - Project Ownership: Partner with a mentor to define and execute a scoped RL project—from initial simulation to deploying a functional demo on humanoid hardware - Sim-to-Real Development: Iterate on learning algorithms in high-fidelity simulators and assist in the 'sim-to-real' transfer process to ensure robust performance on physical robots - Collaborative Engineering: Work alongside our robotics and hardware teams to troubleshoot system-level challenges and understand the interplay between code and motors - Pipeline Optimization: Help refine training pipelines or data processing tools (such as motion retargeting from human demonstrations) to improve how our robots learn - Technical Communication: Present your project findings and hardware results to the broader engineering team, gaining experience in how to translate data into technical milestones Required Qualifications: - A strong theoretical understanding of Reinforcement Learning (RL) and robot dynamics - Proficiency in Python and experience with common RL frameworks (e.g., PyTorch, JAX) - Familiarity with physics simulators such as MuJoCo, IsaacGym, or Drake - Ability to write clean, maintainable code; exposure to C++ is a significant plus - A 'hacker' mentality—you are excited to get your hands dirty, troubleshoot hardware glitches, and see your code move a physical system - Excellent communication skills and a desire to learn from a world-class team of engineers - Currently enrolled in a BS, MS, or PhD program in Robotics, Computer Science, Mechanical Engineering, or a related technical field - Prior experience (academic or personal projects) involving robotic control or machine learning Preferred Qualifications: - Experience with legged robots or robotic manipulators is a plus but not required Required Skills: Reinforcement Learning, Python, Physics Simulators Important Skills: C++