Current
University of Georgia - GAIM Lab
Student Lab Assistant - Deep Learning for Statistical Downscaling
August 2026 - Present · Athens, Georgia
Building reproducible deep-learning systems for high-resolution weather reconstruction.
Computer Science Graduate Student · Researcher & Engineer
Venkata Vivek Panguluri
I am an M.S. Computer Science student at the University of Georgia, building deep-learning research systems for weather data and practical software across security, recommendation, and backend engineering.
About
Research thinking, practical systems engineering
I work on scientific machine learning at UGA's GAIM Lab, focusing on statistical downscaling and high-resolution weather reconstruction.
My broader engineering work spans secure communication systems, explainable recommendation workflows, REST APIs, cloud infrastructure, and full-stack application development.
Selected Work
Systems across ML, security, and full-stack engineering
Three case studies show research discipline, implementation depth, and careful evaluation across different technical domains.
Scientific Machine Learning / Climate Downscaling
Robust Earth Forecast
A research software pipeline for ERA5-to-PRISM temperature downscaling, focused on high-resolution spatial reconstruction, terrain-aware residual modeling, and boundary diagnostics.
Active research codebase; temporal modeling is archived while spatial reconstruction limits are being diagnosed.
Python PyTorch PyTorch Lightning xarray
Cybersecurity / Full-Stack Systems
Adaptive Network Security System
A Java, Spring Boot, and React research system modeling secure communication as an attack-defense-recovery workflow with adaptive security modes.
Public academic/full-stack cybersecurity project; not a third-party-certified production messenger.
Java Spring Boot Spring Security React
Recommendation Systems / Full-Stack
Context-Aware Cross-Domain Recommendation System
A local React and Node.js recommendation prototype connecting food, fitness, and media with contextual signals, feedback history, and explainable adaptive scoring.
Working local full-stack prototype with optional MongoDB and local JSON fallback storage.
JavaScript React Node.js Express
Research
Deep learning for statistical downscaling and high-resolution weather reconstruction
My current work at UGA's GAIM Lab extends deep-learning infrastructure for statistical weather downscaling.
My research tackles a direct but technically demanding problem: converting coarse atmospheric data into higher-resolution spatial fields while keeping experiments reproducible and limitations visible.
Scientific ML Deep learning Statistical downscaling Geospatial data
Statistical Downscaling
My downscaling work maps coarse weather fields into higher-resolution spatial reconstructions with clear validation boundaries.
Geospatial Machine Learning
Gridded weather data, terrain features, and spatial error patterns shape the model design.
Publications
IEEE research output
Verified conference publications in IoT anomaly detection and uncertainty-aware supply-chain decision analysis.
IoT anomaly detection / IoT security
Utilizing Machine Learning Techniques for Detecting Anomalies in IoT Networks
P. Shobha Rani, Mohammad Vajid Ahamed, Kommi Sree Sai Chaithresh, Sakkuru Kundan Srinivas, Panguluri Venkata Vivek
2024 5th International Conference on Recent Trends in Computer Science and Technology (ICRTCST) - April 9, 2024 - Jamshedpur, India - pp. 105-110
IEEE research output from work on machine-learning pipelines for IoT intrusion detection, network telemetry, temporal and behavioral features, protocol-level features, and real-time deployment considerations.
IoT security Anomaly detection Machine learning Network telemetry
Uncertainty-aware supply-chain risk modeling
Evaluation of Supply Chain Design Decisions under Uncertainty – A Case Study Analysis
Joe Prathap P. M., L Sherin Beevi, W Vinil Dani, Panguluri Venkata Vivek, E. Ajith Jubilson
2024 International Conference on Advances in Modern Age Technologies for Health and Engineering Science (AMATHE) - May 16, 2024 - Shivamogga, India - pp. 1-7
IEEE-indexed research output from work on probabilistic simulation, disruption scenarios, demand variability, operational constraints, decision trees, NPV-based modeling, and scenario analysis.
Supply chain analytics Risk modeling Probabilistic simulation Decision analysis
Experience
Research and engineering experience
My current scientific-ML research at the University of Georgia builds on earlier work in IoT security and uncertainty-aware risk modeling.
Student Lab Assistant - Deep Learning for Statistical Downscaling
University of Georgia - GAIM Lab
Contributing to deep-learning research for statistical weather downscaling and high-resolution weather reconstruction.
August 2026 - Present
Athens, Georgia
- Extending an existing PyTorch downscaling framework with a configurable Enhanced U-Net backbone.
- Building Hydra-based model and experiment configuration for controlled architecture studies.
- Implementing PyTorch Lightning checkpoint and training-resume workflows for interrupted research runs.
- Preparing HRRR data workflows for manifest generation, normalization, and paired downscaling experiments.
- Adding pytest regression coverage in a Git, Linux, and HPC-oriented research workflow.
Research Intern - IoT Security and Machine Learning
R.M.D Engineering College
Built machine-learning workflows for IoT intrusion detection, translating network telemetry into anomaly signals for operational monitoring.
2024
Chennai, India
- Designed end-to-end ML pipelines for high-volume IoT network telemetry.
- Engineered temporal, behavioral, and protocol-level features for multi-vector attack behavior.
- Evaluated Random Forest, SVM, and ensemble models with attention to detection accuracy, computational efficiency, and real-time deployment feasibility.
- Contributed to IEEE research output in IoT anomaly detection and security.
Research Intern - Supply Chain Analytics and Risk Modeling
R.M.D Engineering College
Modeled supply-chain behavior under uncertainty using probabilistic simulation, decision trees, and NPV-based risk analysis.
2024
Chennai, India
- Built simulation frameworks for disruption scenarios, demand variability, and operational constraints.
- Developed decision-tree and NPV-based models to compare supply-chain alternatives under financial risk.
- Performed scenario-driven analysis to identify stable, cost-efficient strategies under volatility.
- Co-authored IEEE-indexed research on uncertainty-aware risk modeling for complex supply systems.
Areas of Interest
A broader technical foundation
M.S. Computer Science, University of Georgia. January 2025 - December 2026.
Machine Learning
I work with deep learning, supervised modeling, experiment design, and model evaluation.
Scientific & Geospatial Computing
My scientific work uses weather data, gridded arrays, spatial reconstruction, and scientific ML workflows.
Cybersecurity
Security projects have taken me into IoT security, adaptive defenses, secure communication systems, and attack-defense modeling.
Recommendation Systems
I build context-aware ranking and feedback workflows across food, fitness, and media domains.
Software Systems
My software work includes backend services, REST APIs, frontend workflows, and application architecture.
Cloud & Infrastructure
I use Google Cloud, Docker, Kubernetes, Terraform, and deployment tooling to connect research code with maintainable systems.
Programming
Python Java C SQL JavaScript TypeScript
ML / Data
PyTorch PyTorch Lightning Hydra TensorFlow Scikit-learn Pandas NumPy xarray Herbie
Software Engineering
React Node.js Express Spring Boot Spring Security REST APIs
Cloud / Infrastructure
Git Docker Google Cloud Kubernetes Terraform MongoDB MySQL
Research / Technical Domains
Machine Learning Deep Learning Scientific Computing Geospatial ML Statistical Downscaling Cybersecurity Weather and Climate Modeling Recommendation Systems Supply Chain Analytics
Writing
Working notes and technical essays
I use these notes to write about weather data, experiment discipline, and machine-learning systems.
Technical note
HRRR Data Workflows for Machine Learning
Practical notes on preparing HRRR-style weather data for reproducible machine-learning experiments.
3 min read
Technical note
Reproducible Downscaling Experiments
Notes on experiment boundaries, configuration, and reporting for statistical downscaling work.
3 min read
Contact
Research, engineering, and collaboration inquiries
Based in Athens, Georgia, I use email as the most reliable public contact path.