Applied Scientist Ii, Reinforcement Learning

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

Applied Scientist II role focused on developing advanced robotics systems using AI, deep learning, and reinforcement learning for automation at Amazon's scale. The role involves designing and implementing control methods for balance, locomotion, and manipulation, with a focus on bridging theoretical advancements and practical implementation in robotics.

What you'd actually do

  1. Design and implement whole body control methods for balance, locomotion, and dexterous manipulation
  2. Utilize state-of-the-art in methods in learned and model-based control
  3. Create robust and safe behaviors for different terrains and tasks
  4. Implement real-time controllers with stability guarantees
  5. Collaborate effectively with multi-disciplinary teams to co-design hardware and algorithms for loco-manipulation

Skills

Required

  • PhD or Master's degree with 2+ years of applied research experience
  • Imitation learning
  • Reinforcement learning
  • Whole-body control
  • Learned control methods
  • Model-based control
  • Real-time controllers
  • State estimation
  • Multiple sensor modalities
  • Simulation environments (IsaacLab, Mujoco, Drake, etc.)

Nice to have

  • Java
  • C++
  • Python
  • Cross-functional team collaboration
  • Low-level joint torque/impedance control
  • Teleoperation systems
  • Robotics frameworks (Matlab, ROS)

What the JD emphasized

  • PhD, or Master's degree and 2+ years of applied research experience
  • Experience with imitation learning and reinforcement learning for whole-body control
  • Experience with simulation environments such as IsaacLab, Mujoco, Drake, etc.
  • Experience with developing and deploying code for real-time controllers
  • Experience in state estimation from multiple sensor modalities

Other signals

  • develop advanced robotics systems
  • cutting-edge AI
  • transform automation
  • reinforcement learning
  • deep learning
  • large language models