I am a PhD researcher in Informatics at Indiana University’s Healthy Cities Lab, advised by Dr. Dana Habeeb. My work sits at the intersection of urban climate science, geospatial machine learning, and public health. In simple terms: I study where heat concentrates in cities, who is the most vulnerable, and how we can build practical tools to reduce risk.
Current Projects
Urban Representation Learning for Temperature Prediction
I am developing models to predict hourly 10m near-surface air temperature across entire cities using sparse in-situ sensors. This work uses precomputed Earth observation embeddings from Google’s AEF foundation model to represent urban context, reduce manual feature engineering, and improve transfer across cities.
Local Climate Zone (LCZ) Classification
I built a multi-modal framework to classify urban environments into standardized LCZs using decision tree ensembles and neural network architectures on Google Earth Engine. The system includes end-to-end feature engineering, CNN-based feature extraction, and validation against hyperlocal sensor measurements. Classified imagery on ArcGIS. (Manuscript under review)
Heat-Health Analysis
I use spatiotemporal models to analyze heat-related emergency department visits in two Midwest communities. This work integrates hospital records, weather observations, and census socio-economic data in R to quantify exposure thresholds and demographic vulnerability patterns.
Heat Vulnerability Dashboard
I co-developed a municipal decision-support platform for extreme heat planning. It uses PCA-based index construction across socio-economic, health, and environmental indicators to produce a Heat Vulnerability Index (HVI), delivered through a public-facing interactive web interface. Live dashboard (Richmond, IN)
Urban Sensor Network & Real-Time Dashboard
I helped site and deploy a city-wide distributed temperature sensor network informed by LCZ outputs and urban form analysis. I also built a real-time streaming dashboard (Flask API + Chart.js) to visualize in-situ measurements from across the network. Live dashboard
Publications
Download CV (PDF) · Google Scholar
Journal Articles
Habeeb, D., Devajji, R., et al. (2025). Design and Evaluation of Wearable Solar Radiation Shields for Enhanced Personal Heat Exposure Monitoring. Sensors, 25(3), 945.
Conference Papers & Presentations
Devajji, R., Habeeb, D., & Wilson, J. S. (2025). Extreme Heat and Health Impacts: A Study of Emergency Department Visits in Midwest Communities. American Association of Geographers (AAG) Annual Meeting.
Habeeb, D., Devajji, R., & Wilson, J. (2025). Local Climate Zones Validation Utilizing Hyperlocal Near-Surface Air Temperature Data. American Association of Geographers (AAG) Annual Meeting.
Devajji, R. & Habeeb, D. (2025). A Machine Learning Approach to Predicting Hyperlocal Temperature in the Built Environment. Association of Collegiate Schools of Planning (ACSP) Annual Conference.
Habeeb, D., Devajji, R., Subramanian, L., Davis, L., & Gumaer, J. (2025). Heat Vulnerability Dashboard: A Decision Support Tool for Extreme Heat. Association of Collegiate Schools of Planning (ACSP) Annual Conference.
Tu, H., Devajji, R., & Horan, T. (2025). Algorithmic Literacy and Digital Privacy in the US: An Exploratory Study Using Data Visualization. IEEE International Conference on Advanced Data Visualization (ICAD).
Habeeb, D., Polak, N., & Devajji, R. (2024). Leveraging an Urban Environmental Sensing Network to Improve Extreme Heat Resilience. IEEE International Green and Sustainable Computing Conference (IGSC).
Devajji, R. (2025). SDG 13.3 and Early Warning Systems (EWS): AI in Extreme Heat Resilience. AICTE ATAL Faculty Development Program, RNS Institute of Technology (RNSIT), Bengaluru (3-hour invited session; alma mater).
Tools & Methods
| Category | Tools |
|---|---|
| Remote Sensing | Google Earth Engine, Landsat-8, Sentinel-2, NAIP, GDAL |
| GIS | QGIS, ArcGIS Pro, ArcPy, PostGIS |
| ML / DL | Scikit-learn, PyTorch, Decision Tree Ensembles, CNNs |
| Spatiotemporal | Time-series ML, Spatial interpolation, R (spatial packages) |
| Data & Apps | Python, R, PostgreSQL, Flask, Chart.js, Shiny, Tableau |
| Cloud & Infra | GCP, Docker, Terraform, CI/CD |