About

My research focuses on the hard parts: non-stationary signals, cross-subject and cross-site distribution shift, and models that are fast and reproducible under real-world constraints. I work primarily in PyTorch — transformers, self-supervised and contrastive learning, transfer learning and domain adaptation — backed by rigorous validation.

I’m also deeply engaged with LLMs, multi-agent systems, and AI engineering in practice. And I find teaching genuinely rewarding: turning hard ideas into something a student can actually use.

Contact

Open to collaboration in applied AI, neural engineering, and machine learning. Reach out.