(Senior) AI Engineer - Reinforcement Learning Manipulation

RIVR, part of Amazon is a robotics company pioneering Physical AI through real-world doorstep delivery. Founded in 2024 as an ETH Zurich spin-off, RIVR developed wheeled-legged robots designed to operate in complex, unstructured environments such as stairs, gates, doors, and uneven urban terrain. We believe that achieving general physical intelligence requires solving real customer problems in the real world, where robots can learn from rich operational data at scale.

Following our acquisition by Amazon in March 2026, we are continuing this mission with greater reach and speed. By combining custom robot hardware, onboard autonomy, and cloud-based coordination, RIVR, part of Amazon is building the next generation of safe, reliable autonomous robots for last-mile delivery

Job description: Dexterous Manipulation RL

Reinforcement learning is transforming our robotic intelligence, enabling autonomous behavior without human guidance. We are seeking a Senior AI Engineer with deep expertise in reinforcement learning and deep learning, including supervised and self-supervised learning with a focus on dexterous manipulation. Your role will involve leveraging both simulated and real-world data to address practical challenges in dynamic grasping, contact-rich manipulation, and object interaction. If you are passionate about advancing AI and developing innovative solutions, join us in shaping the future of intelligent robotics.

Develop cutting-edge reinforcement learning algorithms to enable robust, contact-rich dexterous manipulation, translating vision, depth, tactile, and proprioceptive sensor input into precise end-effector and joint-level motor commands.

Design, test, and refine algorithms to solve complex real-world manipulation challenges, such as handling diverse package form factors, dynamic hand-offs, and operating door handles or latches.

Collaborate with the foundation model team to innovate methods that leverage both simulated and real-world data.

Strong background in robotic manipulation, including dynamics, grasp synthesis, and trajectory optimization.

Master’s degree or higher in a relevant field such as Engineering, Robotics, or Machine Learning.

A minimum of five years of industry or research experience, with PhD experience applicable.

Strong deep learning fundamentals, including supervised and self-supervised learning techniques, and reinforcement learning, including Markov Decision Processes (MDPs), neural network architectures, policy optimization algorithms, model-based vs. model-free RL, exploration-exploitation strategies, value function methods, transfer learning, domain adaptation, sim-to-real transfer, etc.

Strong background in robotics including autonomy and/or manipulation.

Experience with deploying artificial neural networks on hardware platforms.

Ability to write production-level code in modern C++.

Ability to prototype algorithms and train deep neural networks in Python.

PhD degree in Robotics, Engineering, Computer Science, Machine Learning or a similar discipline, or an equivalent amount of research experience.

Publications at top-tier conferences (e.g., ICRA, IROS, CoRL, RSS) specifically focusing on robotic manipulation, grasping, or contact-rich RL.

Demonstrated experience working with tactile sensing, multi-fingered robotic hands, or bimanual manipulation.