Nafiz Imtiaz Khan
Hi! I am pursuing PhD in Computer Science at University of California - Davis, where I work in the CS AI4Health Lab and DECAL lab with professor Vladimir Filkov. I am co-advised by Professor Roger Eric Goldman. I also serve as Associate Lab Manager at the AI for Health Lab. Formerly, I served as a Lecturer in the Department of Computer Science and Engineering (CSE) at MIST.
My research applies large language models, vision-language models, and machine learning to software engineering and health informatics. In health, I build vision-language models for interventional radiology that learn from angiography videos and the reports radiologists already write, so that clinical knowledge can be captured without costly manual annotation. In software engineering, I study how LLMs can ground their answers in real evidence, through tools such as EvidenceBot and OSSPREY that help developers and researchers reason about software projects. Across both areas, my recent work covers retrieval-augmented generation (RAG), fine-tuning LLMs for domain-specific tasks, careful evaluation of what these models actually understand, and building scalable ML pipelines for real-world problems.
Shoot me an email at nikhan@ucdavis.edu.
News
- Oct 2026 Presented MedVision at the UC Davis Radiology Research Symposium.
- Aug 2026 RepoWise won the Best Paper Award at SEET (Springer CCIS).
- Jun 2026 Started a summer internship at BetterHelp, working on reducing bias in language models.
- Nov 2025 OSSPREY published at the ASE 2025 Tool Demonstrations track.
- Aug 2025 Joined Scale AI as a Human Frontier Collective Specialist (GenAI).
- Jun 2025 EvidenceBot published at the FSE 2025 Tool Demonstrations track.
Education
-
PhD in CS (2025 - Current)
University of California - Davis
-
MSc in CS (2023 - 2025)
University of California - Davis
-
BSc in CSE (2017 - 2021)
Military Institute of Science and Technology (MIST)
Last updated: —