Rohan Singh Wilkho: AI for managed water systems, Cornell University
About Me
I am an Eric and Wendy Schmidt AI in Science Postdoctoral Fellow in the School of Civil and Environmental Engineering at Cornell University, working with Prof. Stefano Galelli. I use artificial intelligence and machine learning to study managed water systems (dams, canals, and diversions) that, along with hydroclimatic forcing, shape water flow in our rivers.
Before joining Cornell, I was a Postdoctoral Researcher and Visiting Lecturer in Civil Engineering at Texas A&M University, where I earned a PhD in Civil Engineering and a Master of Computer Science, both in 2024, advised by Prof. Nasir Gharaibeh. Earlier, I worked as a tunnel engineer on India’s first underwater tunnel project (2016–2018), after earning a bachelor’s in Civil Engineering from Jadavpur University in 2016.
Research Program
Most of the world’s major river basins are managed systems, operated by human judgment that hydrologic models cannot see and operating manuals do not record. My research program makes AI and water systems engineering inform each other: the coupling between human decisions and physical outcomes exposes limitations in predictive modeling, which cannot capture causality and human intent. These limitations motivate methodological innovations in inverse reinforcement learning, physics-informed AI, and multi-modal AI, which provide the foundation to characterize continuously evolving human-water interactions. The central goal of my research is to understand how the decision logic of managed water system operators must evolve to sustain water security and flood control under hydroclimatic non-stationarity and a changing climate.
My research program is built on three thrusts:
- Decoding the behavioral logic of water system operators, to recover the objectives that actually govern their decisions.
- Reconstructing community-scale performance records, to reveal what communities experience when those decisions fail.
- Diagnosing performance under compound pressures, to identify how operating logic must evolve to sustain design performance.
My AI methods and expertise have also extended to other civil engineering domains, with published studies in disaster resilience (PRIME, DFFS, FlowsDT) and infrastructure condition assessment (CulvertVision, Satellite-to-Street).
Supported by: Schmidt Sciences, US Army Corps of Engineers, Texas Department of Transportation, and National Science Foundation.
Service: (1) Peer reviewer (44 reviews, 20 journals); (2) Editorial team, Sustainable Cities and Society and Environmental Management and Sustainable Development.
Mentorship: Co-mentored nine graduate students (Ph.D., M.S., and M.Eng.) at Cornell University and Texas A&M University.
Selected Publications
Full publication list on 
Google Scholar.
| Water Resources Research |
Decoding decision-making in managed hydrologic systems with inverse reinforcement learning (Under Review) |
|---|---|
| ASCE Journal of Computing in Civil Engineering |
Satellite-to-Street: A Generative Deep Learning Framework for Pavement Condition Assessment from Satellite Imagery (Under Review) |
| ASCE Journal of Infrastructure Systems |
CulvertVision: Advancing AI-Augmented Inspection and Condition Assessment of Culverts |
| International Journal of Disaster Risk Reduction | |
| Advanced Engineering Informatics |
FF-BERT: A BERT-based ensemble for automated classification of web-based text on flash flood events |
| Environmental Modelling & Software |
Funded Research Grants
| Empire AI (2026) |
Learning and Simulating Human Actions via Inverse Reinforcement Learning for Resilient River Basin Management Role: Co-PI PI: Stefano Galelli, Cornell University |
|---|---|
| NSF ACCESS (2025) |
Learning and Simulating Human Actions via Inverse Reinforcement Learning for Sustainable River Basin Management Role: Co-PI PI: Stefano Galelli, Cornell University |
| FHWA Accelerating Market Readiness (2025) |
Integrated Remote Sensing and AI for Pavement Condition Monitoring Role: Co-PI, Lead Developer PI: Nasir G. Gharaibeh, Texas A&M University |
| Schmidt AI in Science Postdoctoral Fellowship (2025) |
Learning and Simulating Human Actions via Inverse Reinforcement Learning for Resilient River Basin Management Role: PI |
| NSF #1931301 (2019) |
Reducing the Human Impacts of Flash Floods: Development of Microdata and Causal Model to Inform Mitigation and Preparedness Role: Co-author, Participant, Lead Developer PI: Nasir G. Gharaibeh, Texas A&M University |
Teaching Experience
| 2025 |
Visiting Lecturer (Instructor-of-record) CVEN 303: Civil Engineering Measurement |
|---|---|
| 2022–2024 |
Graduate Teaching Assistant |
Selected Invited Talks & Workshops
| September 2026 |
“Inverse Reinforcement Learning” CS 6703 (AI for Science) |
|---|---|
| June 2026 |
AI for Multimodal Science (AIMS) 2026 |
| June 2026 |
“Modeling the Operations of Reservoir Systems with LLMs and Inverse Reinforcement Learning” MultiSector Dynamics |
| April 2026 |
NOAA Workshop on Advancing Social and Physical Science Data to Reduce Flood Losses Natural Hazards Center |
| April 2026 |
“Stochastic Model Predictive Control” CEE 6400 (Artificial Intelligence for Control of Environmental Systems) |
| February 2026 | |
| December 2025 |
“Cognitive Civil Systems: From Multimodal Sensing to Behavioral Sensing” Department of Civil Engineering |
Selected Milestones
| August 2026 | Lead-author paper by mentee Xinke (Ada) Huang, “Constraint-Based DEM Enhancement Using LiDAR Data at Sparse Highway-Bridge Locations,” published in ASCE Journal of Surveying Engineering |
|---|---|
| July 2026 | “CulvertVision: Advancing AI-Augmented Inspection and Condition Assessment of Culverts” accepted in ASCE Journal of Infrastructure Systems |
| February 2026 | Secured Empire AI computing grant for GPU compute resources |
| February 2026 | “FlowsDT: A Geospatial Digital Twin for Navigating Urban Flood Dynamics” accepted in Computers, Environment and Urban Systems |
| January 2026 | Selected to the Early Career Editorial Board, Sustainable Cities and Society |
| November 2025 | Secured NSF ACCESS computing grant for GPU compute resources |
| November 2025 | Road-Lens (InfraFrame) awarded FHWA Accelerating Market Readiness grant |
| October 2025 | Began Schmidt AI in Science Postdoctoral Fellowship, Cornell University |
| March 2025 | “FF-NER: A named entity recognition model for harvesting web-based information about flash floods and related infrastructure impacts” accepted in International Journal of Disaster Risk Reduction |
| November 2024 | “A GIS-based tool for dynamic assessment of community susceptibility to flash flooding” accepted in Journal of Flood Risk Management |
| October 2024 | “PRIME: A CyberGIS Platform for Resilience Inference Measurement and Enhancement” accepted in Computers, Environment and Urban Systems |
| August 2024 | PhD in Civil Engineering, Texas A&M University |
| May 2024 | Master of Computer Science, Texas A&M University |
| April 2024 | 1st runner-up, 3MT Thesis Competition, Zachry Department of Civil and Environmental Engineering, Texas A&M University |
| December 2023 | “FF-BERT: A BERT-based ensemble for automated classification of web-based text on flash flood events” accepted in Advanced Engineering Informatics |
| September 2023 | “Predicting Flash Flood Economic Damage at the Community Scale: Empirical Zero-Inflated Model with Semicontinuous Data” accepted in ASCE Natural Hazards Review |
| June 2023 | “FF-IR: An information retrieval system for flash flood events developed by integrating public-domain data and machine learning” accepted in Environmental Modelling & Software |
| May 2023 | “Environmental, climatic, and situational factors influencing the probability of fatality or injury occurrence in flash flooding: a rare event logistic regression predictive model” accepted in Natural Hazards |
| May 2023 | 1st runner-up, Association of State Flood Plain Managers 2023 Student Paper Competition |
| April 2023 | Montgomery Award (Texas A&M University’s university-wide honor for graduate student leadership and scholarship, selected from a highly competitive pool), Graduate and Professional School |
Collaboration: Interested in talking, sharing ideas, or collaborating on AI for water and infrastructure? Reach out to me. Please use the subject line “AI for water/infrastructure.”