Principal Applied Scientist, Robotics

Amazon Amazon · Big Tech · N.reading, MA · Applied Science

This role focuses on developing advanced robotics systems that integrate AI, control systems, and mechanical design for automation. The scientist will define the scientific roadmap for whole body control and dexterous manipulation, applying deep learning and LLMs to solve complex operational challenges in dynamic environments. The role involves research and practical implementation of AI in physical robotic hardware, with a focus on shipping these systems.

What you'd actually do

  1. Define and drive the long-term scientific roadmap for whole body control and dexterous manipulation, working with autonomy and delivering artifacts that set the standard for scientific and engineering excellence
  2. Serve as the primary technical authority on whole body control methods — including reinforcement learning, imitation learning, hierarchical quadratic programming, and model-predictive control — across the organization
  3. Identify and tackle intrinsically hard, open-ended research problems in loco-manipulation, acquiring expertise as needed and proposing innovative solutions that span multiple teams
  4. Collaborate with hardware and robotics leads to co-design systems for loco-manipulation, ensuring science solutions are grounded in real-world deployment constraints
  5. Represent scientific capabilities to senior leadership and external partners; communicate complex technical concepts to both technical and non-technical audiences

Skills

Required

  • whole body control methods
  • hierarchical quadratic programming (HQP)
  • model-predictive control (MPC)
  • imitation learning
  • reinforcement learning
  • real-time controllers on physical robotic hardware
  • simulation environments (IsaacLab, MuJoCo, Drake)
  • state estimation from multiple sensor modalities
  • technical strategy influence

Nice to have

  • co-designing hardware and algorithms for loco-manipulation systems
  • mentoring scientists and engineers
  • PhD in Robotics with a focus on whole body control or dexterous manipulation
  • publications and/or patents in robotics, control, or machine learning

What the JD emphasized

  • PhD in Robotics, Computer Science, Mechanical Engineering, or a related field, with 7+ years of relevant research experience after degree; or Master's degree with 12+ years of equivalent experience
  • Deep expertise in whole body control methods, including hierarchical quadratic programming (HQP) and model-predictive control (MPC)
  • Proven experience with imitation learning and reinforcement learning applied to whole body control and manipulation
  • Experience developing and deploying real-time controllers on physical robotic hardware

Other signals

  • advanced robotics systems
  • cutting-edge AI
  • deep learning
  • large language models
  • reinforcement learning
  • imitation learning