Hi, I'm Deepayan Basu
A passionate B.Tech student at IIT Jodhpur, specializing in Materials Science. I'm driven by innovation and eager to contribute to the tech world. Explore my portfolio to see my projects and skills.
About Me
I'm a senior undergraduate from IIT Jodhpur specializing in Materials Science and Metallurgy. I’ve worked on predictive ML models for material properties and developed NLP and LLM-based financial analytics tools. I'm passionate about data science, deep learning, and interdisciplinary innovation.
Projects
SAR Target Classification (J-CNN + TDP-SAR)
Developed a robust, frequency-domain optimized CNN for Synthetic Aperture Radar (SAR) military target recognition.
Learn More →Band Gap Prediction Using DFT + ML
Created a predictive model for material band gaps using data from the Materials Project API with VASP and Python.
Learn More →LLM-Based Financial Chatbot
Built an AI assistant to extract, summarize, and search insights from 100+ financial transcripts using LangChain, Ollama, and embeddings.
Learn More →Lankford coefficient Prediction for Crystal Anisotropy
This project employs a Convolutional Neural Network (CNN) to predict Lankford coefficients (r-values) based on Orientation Distribution Function (ODF) images.
Learn More →Predicting Stress Hotspots in FCC Materials
The project leverages microstructural features from both texture-derived (e.g., misorientation, slip transmission) and geometry-derived (e.g., grain shape, size) features to predict stress hotspots in FCC materials.
Learn More →Causal Machine Learning analysis to boost user engagement in App participation
Applied Causal Random Forests and Causal Tree models on survey data (604 participants) to analyze Conditional Average Treatment Effect (CATE) by age, revealing key drivers of app engagement across demographics.
Learn More →Skills
Python
SQL
JavaScript
React
Streamlit
TensorFlow
LangChain
VASP
LAMMPS
NLP