Summary
Colin Kohler is a roboticist and PhD candidate in computer science at Northeastern University with 11 years of software and robotics experience focused on perception-to-planning pipelines for dexterous manipulation. He combines deep learning (supervised, semi-supervised, and reinforcement learning) with novel action abstractions that encode motion primitives directly into depth-camera pixel spaces to enable pick, place, and push behaviors. His work bridges rigorous academic research with applied robotics internships and industry roles, most recently at the RAI Institute after multi-year research at Northeastern and internships at TRACLabs and The AI Institute. Colin’s research emphasis on encoding motion primitives in observation space is a practical, less obvious lever that improves sample efficiency and robustness in manipulation. Based in Boston, he maintains an active GitHub and Google Scholar presence, signaling a strong record of reproducible research and open engineering.
10 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Northeastern University
English