Harry Papadakis

Hi! I’m Harry — an incoming PhD student at the MIT Operations Research Center, starting Fall 2026 under Prof. Dimitris Bertsimas. My given name is Charidimos, which is what you’ll see on my papers; Harry is the day-to-day version.

I recently completed my MEng in Electrical & Computer Engineering at the National Technical University of Athens, where I wrote my diploma thesis with Prof. Giorgos Stamou on adaptive multi-agent LLM systems for financial trading. I currently work as an AI engineer building agentic LLM systems for the legal domain — enabling contract review, clause-level risk identification, and collaborative AI-assisted editing inside lawyers’ native workflows.

My research interests sit at the intersection of operations research and machine learning. On the applied side, I’m drawn to combining optimization and ML to tackle high-impact, real-world problems — particularly in maritime operations, where scale, uncertainty, and data richness make it a natural meeting point for the two fields. On the theoretical side, I’m interested in the deeper interplay between AI and operations research — and increasingly in whether LLMs are quietly changing what kinds of mathematical and algorithmic discovery are possible. Recent results, like LLMs producing new constructions for long-open Erdős problems, hint at something stronger than automation: models that can propose conjectures, find structures beyond human intuition, and shape which questions are worth asking.

Feel free to reach out — contact links are in the sidebar.