Research

Themes I'm working on now. Each is a curated narrative, not a publication dump.

  • Computer vision in healthcare

    Robust segmentation and classification on small, heterogeneous medical datasets — where domain shift from training to deployment is the hard problem.

    computer-visionmedical-imaginghealthcare

    Continue reading →

  • LLMs for translational medicine

    Using large language models as interfaces to structured medical knowledge — where the work lives in retrieval, constraints, and ground-truth signal, not the model itself.

    large-language-modelsnatural-language-processinghealthcaregenomics

    Continue reading →

  • Multi-agent systems for robotics

    Composing perception, planning, and control modules from heterogeneous model families into robotic stacks that are reconfigurable, observable, and tolerant of partial failure.

    multi-agent-systemsroboticslarge-language-models

    Continue reading →

  • Medical devices and embedded AI

    On-device inference under real power, latency, and clinical-environment constraints — closing the gap between papers and devices.

    medical-devicesembedded-aiedge-aimodel-compression

    Continue reading →