cv

B.Tech

National Institute of Technology, Hamirpur

Electrical Engineering

HSC

Yashoda High School and Junior College

Higher Secondary Education

SSC

Somalwar High School, Ramdaspeth

Secondary School Certificate

Predoc Associate

• Developed comprehensive Machine Learning systems for CASDIC's combat aircraft applications
• Evaluated fine-tuned IBM Granite & Gemma models for SciML using a developed Graph-RAG pipeline, achieving SOTA accuracy on benchmark datasets
• Lead Open Source Projects PULSE & SciREX at AiREX lab under Professor Sashikumaar Ganesan

Data Scientist

• Designed and implemented a high-performance statistical model for fraudulent transaction prediction, achieving the highest abuse detection rate among all models
• Spearheaded the end-to-end development of a fraud detection pipeline using graph database technologies (Memgraph, Neo4j, Cypher), after evaluating 10+ platforms

Machine Learning Intern

• Developed ML models predicting stock prices using Python, achieving 85% accuracy. Incorporated news and market data to improve forecasts
• Implemented time series analysis and feature engineering, boosting model accuracy by 25% & enabling data-driven decision making

Open Source Technical Writer

• Collaborated with the writing team to improve existing user documentation and beginner tutorials in differential privacy
• Active technical writer and contributor, publishing blog posts and articles about federated learning and privacy-preserving machine learning resulting in increased product adoption

Document Tampering Detection

• Developed a comprehensive forgery detection system with YOLOV8 to identify and classify fraudulent alterations such as overwriting, scribbling, and whitener use
• Built dataset of 600+ document images for training
• Designed a Streamlit-based web interface facilitating intuitive document uploads, enabling users to visualize tampering and analyze detailed results

Fundus AI

• Developed complete solution to detect eye and systemic diseases such as Diabetic Retinopathy, Cataract, Glaucoma, ARMD, Hypertensive Retinopathy
• Developed deep learning smartphone app for efficient ophthalmic disease screening utilizing device camera for accessible healthcare
• Designed 3D-printed smartphone accessory using Tinkercad for capturing high-quality fundus imaging compatible with all smartphones
• Technologies used: Python, Keras, Tensorflow, Flutter, Tinkercad