Ananya A. Joshi
I'm an Assistant Professor in Psychiatry & Behavioral Science with a secondary appointment in Computer Science at Johns Hopkins University. My work is about computational and Generative AI methods and systems for diagnosing, treating, and monitoring mood disorders. I'm also broadly interested in computational and data science systems with clinical applications.
Current Research Themes
- • GenAI Guardrailing and Monitoring: Methods and systems with probabilistic guarentees for detecting and monitoring mood disorders.
- • Drug Matching: Online, high-dimensional statistical matching from noisy and missing data.
- • Experimental Design: Large-scale digital clinical interventions and feedback loops for evaluating GenAI tools in clinical psychiatry.
Opportunities for Students
I'm excited to advise students on these multidisciplinary topics. If you're a Hopkins student and excited about this research, let's chat!
Background
I received my Ph.D. and M.S. in Computer Science from Carnegie Mellon University, supported by the NSF Graduate Research Fellowship Program, and my B.S.E. in Computer Science from Princeton University. During my Ph.D., I developed AI methods for data monitoring while simultaneously serving as a project manager to deploy this research in an applied public health setting that's now been used for over 2 years. I've also spent time as a software engineer and researcher in networked and operating systems, protein folding, and general large language model (LLM) alignment — work that has been honored with Rising Stars and Best Paper Awards. Outside of academia, my perspectives toward research and impact have been shaped by my Fulbright year in Singapore, an internship with IBM's Human-Centered AI team in Nairobi, a semester abroad at ETH Zurich, and service that was recognized through a Graduate Student Service Award.
Like many, I’ve seen how mood disorders impact the lives of people we love. Being in the School of Medicine, I'll be researching the computational questions that can directly help clinicians and patients improve outcomes.
If you’re interested in learning more, please contact me at aa < lastname > at andrew dot cmu dot edu.