Amin Atrash

About Amin Atrash

I’m a machine learning leader and roboticist with a Ph.D. from McGill University, focused on building intelligent systems that operate robustly in the real world. My work spans probabilistic modeling, reinforcement learning, inverse reinforcement learning, and large language models, with a consistent emphasis on deploying these methods in user-facing systems. I’ve led and contributed to the development of autonomous robots and AI-driven products at Amazon Lab126 (including the home robot Astro) and SRI International, and I am currently a Research Scientist at Rational Dynamics, where I build agentic AI systems for financial applications. Across academia and industry, my work bridges advanced learning and decision-making algorithms with the practical constraints of real-world deployment, spanning human-robot interaction, state estimation, control, speech recognition, and now autonomous financial agent, to deliver intelligent systems that are both technically rigorous and production-ready.

Experience

1. Research Scientist

   Rational Dynamics

   Nov 2025 — Present

Education

• Georgia Institute of Technology

  Bachelor of Science (BS), Computer Science

  1996 — 1999

• McGill University

  Doctor of Philosophy (PhD), Computer Science

  2005 — 2010

• Georgia Institute of Technology

  Master of Science (MS), Computer Science

  1999 — 2003

Skills

• Hidden Markov Models
• Ubiquitous Computing
• Intelligent Systems
• Probabilistic Models
• Human Computer Interaction
• Markov Decision Processes
• Speech Recognition
• Natural Language Processing
• Mobile Robotics
• Robotics
• Machine Learning
• Assistive Technology
• Artificial Intelligence