Auditable AI Systems for Biomedicine

I am a Postdoctoral Researcher at Stanford University, working with Prof. James Zou and Prof. Lei Xing. I earned my PhD in Data Science from New York University, where I was fortunate to work with Carlos Fernandez-Granda, Narges Razavian, and Kyunghyun Cho.

My research builds reliable and auditable AI systems that can reason, adapt, and act in complex real-world environments. I develop methods for robust learning, controllable inference, and agentic AI, with a particular focus on biomedicine—where I am interested in turning heterogeneous evidence into trustworthy decisions, scientific discoveries, and ultimately better paths to treatment.

Outside of academia, I play tennis and am a certified scuba diver and surfer. Some of that lives on the gallery page.

Latest News

View all
  • 2025.10

    Organizing a workshop on AI and healthcare at ICCV 2025 in Hawaii.

  • 2025.09

    Co-authored work on a trainable multi-agent system is online — over 900 stars on GitHub.

  • 2025.09

    Work on unsupervised manifold learning from dynamic brain data accepted at Nature Computational Science.

  • 2025.09

    Paper on reasoning and hallucination in MLLMs accepted at NeurIPS 2025.

  • 2025.09

    Invited talk at the Symposium on AI Medicine, Stanford University.

From Autopilot to Copilot

Robustness

Handle noisy, imperfect data, and bias — learning that stays reliable when the data is not.

Steerability

Adapt to feedback and evolving context at inference time, without retraining.

Agency

Reason and act on complex, real-world tasks as a collaborative partner.

Interested in collaboration?

We are actively seeking students and collaborators with a background in machine learning, large foundation models, AI agents, or AI for medicine.

Send an Inquiry