Overview In this role you will bring learned manipulation policies to life on real robots within Amazon Robotics. You will work with cross-functional teams to close the loop between simulation and hardware, making non-prehensile, contact-rich behaviors fast, robust, and scalable. You'll deploy policies on diverse, demanding hardware and drive improvements to sensing, rewards, and control. This position offers the opportunity to publish and contribute to academic dialogue while delivering impactful robotics capabilities at scale.
Compensation / Benefits- Medical, Dental, and Vision Coverage
- Maternity and Parental Leave Options
- Paid Time Off (PTO)
- 401(k) Plan
- Sign-on payments and RSUs
- Comprehensive health benefits and life insurance
Responsibilities- Deploy and evaluate learned manipulation policies on physical robots under diverse, demanding conditions
- Diagnose gaps between simulation and real-world behavior and adjust policies, rewards, sensing, or control accordingly
- Develop tactile, actuation, and contact-aware methods for fast, robust non-prehensile manipulation
- Write production-quality, real-time code for robot systems
- Build data collection, logging, and evaluation loops to convert real-world data into policy improvements
- Collaborate with control, perception, and hardware teams to move ideas from prototype to hardware demonstration
- Represent Amazon Robotics in academia through publications and talks
Key requirements- 3+ years of building models for business applications (or equivalent)
- PhD, or Master's degree plus 4+ years of science/engineering experience
- Experience publishing in top conferences or journals
- Proficiency in Java, C++, Python or related languages
- Hands-on experience deploying and evaluating learning-based control or manipulation on physical robots
- Strong background in reinforcement learning, learning-based manipulation, contact-rich control, or force/torque control
- reinforcement learning
- learning-based manipulation
- contact-rich control
- force/torque control
- tactile sensing
- compliant control
