I'm an AI/ML engineer who bridges the gap between research and production. Third-year B.Tech student at AISSMS IOIT (SPPU), building systems with LLMs, RAG architectures, and deployed ML models.
Currently open to collaborations in GenAI, RAG systems, and applied deep learning research.
Architected and developed a high-performance interactive website for the college's environmental initiative. Leveraged Next.js 16 and React 19, integrated immersive 3D elements using React Three Fiber, and implemented complex animations via GSAP and Framer Motion. Features include AI-powered interactions using the Google AI SDK.
Monte Carlo financial simulation engine with 500+ stochastic paths across multi-asset Indian portfolios. Engineered a 7-node LangGraph + Gemini pipeline generating structured AI reports with Chart.js visualizations using live yfinance/Prophet market data.
AI-powered teaching assistant for any subject integrating PDF textbooks and YouTube lectures into a seamless RAG pipeline.
Model-Context-Protocol server designed to be called by external AI agents for email composition and calendar management.
Full-stack AI diagnostic system with online retraining, MLflow experiment tracking, and a human-in-the-loop learning pipeline.
A fully functional interpreter written in Go — exploring language theory and compiler design from the ground up.
AI-Powered NFT minter on the Sui Blockchain — bridging generative AI with decentralised asset creation.
Comparative study of DenseNet, VGG, and ConvNeXt — achieving 96% accuracy with ConvNeXt.
Explored Vision Transformers for deepfake detection — 87.33% accuracy on 5,000 images.
Developed a sophisticated AI-driven financial decision simulation engine for the Fintech category during this intensive 8-hour hackathon.
Won the GitHub Track at HackByte 4.0, a Major League Hacking sanctioned hackathon.
Built BugChan — a decentralised bug-bounty platform at the ETH Online hackathon, placing 2nd.
Led end-to-end planning for a conference with 90+ delegates and 110+ volunteers.
Managed interactions and committee agendas for over 90 delegates across multiple committees.
Open to collaborations in GenAI, RAG systems, and deep learning research. Whether it's a project, role, or just a conversation — reach out.