Embedded Files
❖ Research Project (Asst.) | ISRO - IIRS: Dept of Space, Gov of India
(Location: Dehradun, Uttarakhand)
(Feb 10 – June 30, 2025)
Gained expertise in handling geospatial data & techniques to analyze urban microclimate analysis through ENVI-met software, dashboard & data visualization. Second objective – ML & deep neural network-based thermal comfort analysis.
Worked on the research paper titled "Microclimate and Thermal Comfort Modelling at the Neighbourhood Scale: A Case Study of Dehradun".
Developed and integrated thermal remote sensing-based urban microclimate analysis using simulated and ground-based observations, ground-truth data collected via field visit-deployed Automated Weather Stations (AWS device)
Frameworks/Skills: ENVI-met, RayMan, QGIS, SketchUp, Microclimate Modeling, Natural Sciences.
Supervisor (s): Scientist/Engineer - Dr. Hina Pande (Scientist - 'SG', PRSD dept and HOD, hina@iirs.gov.in), Dr. Poonam S. Tiwari (Scientist - 'SG', GITDL dept, poonam@iirs.gov.in), and Dr. Asfa Siddiqui (Scientist - 'SF', URSD dept, asfa@iirs.gov.in)
❖ Project Trainee (Intern) | Indian Space Research Organization (ISRO) Meghalaya: NESAC
(Location: Shillong)
(Dec 2 – Jan 27, 2025)
Lightweight transformer-based visual question answering in remote sensing. Dataset used: LISS4 Indian satellite imagery (.tif format – especially the north-east region of India), trained on 210 images with 3 bands (Red, Blue, NIR), Software: QGIS for geospatial insight from data, applied: BERT, ResNet50 using PyTorch.
Build a lightweight transformer-based Visual Question Answering (VQA) system for remote sensing applications.
Utilized LISS-IV Indian satellite imagery data (.tif) focused on the North-East region of India and trained models on 210 multi-band images (Red, Blue, NIR).
Implemented deep learning models using BERT for question encoding and ResNet-50 for visual feature extraction with PyTorch.
Frameworks/Skills: PyTorch, BERT, ResNet-50, Transformer Models, LISS-IV, Satellite Data Analysis.
Guided by (Scientist/Engineer ‘SE’): Dr. Avinaash Kr Chauhan (GID dept, avinash.chouhan@nesac.gov.in)
❖ Research Internship | Indian Institute of Information Technology, Guwahati (IIITG)
(Location: Assam)
(May 18 – July 18, 2024)
(Application domain: Finance Economics)
Developed short-term stock market forecasting models using machine learning algorithms on 22 years of TCS stock data from Yahoo Finance, achieving 86.2% accuracy.
Experimented with deep learning architectures, including RNN and LSTM, and fine-tuned models using Random SearchCV for hyperparameter optimization.
Frameworks/Skills: Scikit-learn, TensorFlow, LSTM (Bidirectional and Stacked variants), Time Series Analysis.
Supervisor: Dr. Hari K Chaudhury (Associate Professor of Economics, hari@iiitg.ac.in)
❖ HackGPT Celesta ’23 Hackathon
Indian Institute of Technology (IIT) Patna
Timeline: Nov 4 – Nov 5, 2023
Participated in the HackGPT Celesta ’23 Hackathon, contributing to the development and deployment of a voice-to-text chatbot using HTML, CSS, JavaScript, and Node.js. Enhanced full-stack development and problem-solving skills through this project.
Page updated
Google Sites
Report abuse