I am a Senior Reinforcement Learning Engineer at Quilter AI, where I build generative and reinforcement learning systems for physics-constrained PCB design. My work spans generative modeling and online, offline, and multi-agent reinforcement learning, with a focus on scalable systems for complex, constrained optimization problems.

I completed my PhD in Computer Science at the University of California, Irvine. My research spans multi-agent reinforcement learning, game theory, optimization, and generalization.

Previously, I worked on:

• Contextual-bandit methods for in-context learning at Microsoft Research as a Reinforcement Learning Open Source Fest fellow
• Reinforcement learning for token-level LLM hallucination control at Qualcomm AI
• Multi-agent reinforcement learning for combinatorial problems at Zebra Technologies with Biswa Sengupta
• Privacy-preserving reinforcement learning at the UCI Networking Group
• Reinforcement learning for industrial power control, advised by Qiuhua Huang at Pacific Northwest National Laboratory
• Music information retrieval at the Language & Speech Processing Institute of the Athena Research and Innovation Center

I also hold an MS in Computer Engineering from the University of California, Irvine, and an MEng Diploma in Electrical and Computer Engineering from the National Technical University of Athens. For more information, see my CV or contact me.

I am originally from Rethymno and grew up in Athens, Greece. If you are planning a visit to Crete, ask me for recommendations.