Rohan Pandey

PhD Student · ML & NLP for Science & Healthcare · UMass Amherst

I'm a fourth-year PhD student in the Bio-NLP Lab at UMass Amherst, advised by Hong Yu. In a previous life, I helped develop pre-ChatGPT-era AI platforms for spend analytics at McKinsey (Source AI and Spendscape).

I work on understanding and building AI that holds up when it meets the real world—especially in science and healthcare.

I spend most of my time thinking about:

  • AI for science: how AI can be genuinely useful to scientists, for example helping them find the lesser-known prior work that sparks new ideas (MUSES), and what widespread AI use means for how science is done, validated, and communicated.
  • AI for healthcare: using AI to tackle important public-health challenges, like homelessness and opioid overdose, and understanding what it takes to make AI work in real-world healthcare settings, from messy electronic health records (EHRs) to clinical workflows.
  • Science of AI: understanding what makes AI systems, especially large language models (LLMs), work (or fail), and how we can make them better.

I'm looking for research scientist internships for Summer 2027, especially in LLMs, AI for science, or healthcare. I'm always open to interesting research conversations and collaborations, and I enjoy mentoring students and early-career professionals navigating their academic and career paths. I've been fortunate to receive a lot of support and guidance along the way, and I hope to pay that forward. If you'd like to collaborate, seek advice, or simply connect, feel free to reach out!

Photo of Rohan Pandey

Publications

  1. OverdoseMoE: A Multi-Expert Framework for Opioid Overdose Risk Prediction

    Mingchen Li, Rohan Pandey, Junhui Qian, Feiyun Ouyang, Sunjae Kwon, Hong Yu

    arXiv preprint, 2026 Under review·arXiv·PDF·

  2. MUSES: A Benchmark for Prospective Intellectual-Roots Retrieval

    Rohan Pandey, Sunjae Kwon, Hong Yu

    arXiv preprint, 2026 Under review·arXiv·PDF·

  3. Predicting first-episode homelessness among US Veterans using longitudinal EHR data: time-varying models and social risk factors

    Rohan Pandey*, Haijuan Yan, Hong Yu, Jack Tsai

    arXiv preprint, 2026 Under review·arXiv·PDF·

  4. A machine learning application for raising WASH awareness in the times of COVID-19 pandemic

    Rohan Pandey*, Vaibhav Gautam, Ridam Pal, Harsh Bandhey, Tavpritesh Sethi

    Nature Scientific Reports, 2022·Paper·Code·

  5. COVID-19 Mask Usage and Social Distancing in Social Media Images: Large-scale Deep Learning Analysis

    Asmit Kumar Singh*, Paras Mehan*, Divyanshu Sharma*, Rohan Pandey*, Tavpritesh Sethi, Ponnurangam Kumaraguru

    JMIR Public Health and Surveillance, 2022·Paper·

  6. COVIDScreen: Explainable deep learning framework for differential diagnosis of COVID-19 using chest X-Rays

    Rajeev Kumar Singh*, Rohan Pandey*, Rishie Nandhan Babu*

    Neural Computing and Applications, 2021·Paper·

  7. Cervical Cancer Diagnosis using CervixNet — A Deep Learning Approach

    Rohan Gorantla*, Rajeev Kumar Singh*, Rohan Pandey*, Mayank Jain*

    IEEE BIBE (Bioinformatics and Bioengineering), 2019·Paper·

Experience