Academic Journey
Timeline
Cardiff University
MSc in Advance Computer Science with placement Year
2025 - Now
National Institute of Technology, Kurukshetra, India
B.Tech in Computer Engineering
2020 - 2024
- Got My Undergradute Degree
My Work
RiskLensAI
RiskLensAI is an explainable AI credit-risk decision support system that predicts default probability using calibrated ML models and real financial datasets. It combines an ML pipeline, SHAP-based explainability, and a full-stack dashboard to help analysts make transparent, threshold-driven lending decisions.
Irfan Express
AI travel itinerary generator built with Next.js + Gemini API, producing day-wise plans with activities, travel time, and budget notes. Includes user profiles, saved trips, and editable itinerary drafts backed by PostgreSQL (Neon) + Drizzle.
Sahina Beauté — Full Stack Website for a Beauty Salon (Client Work)
Designed and developed a production-ready website for a women-only Indian beauty salon in Saint-Denis, France. The platform includes a service catalogue, pricing menu, appointment booking form, and automated email confirmations using Resend. Built with Next.js and deployed on Vercel for fast global delivery.
Ripple (Ongoing) — Personalized Feed Ranking Engine
A ML-powered content ranking and recommendation engine that personalizes user feeds using interaction signals and learning-to-rank models. Designed to simulate real-world social feed ordering with real engagement datasets and evaluation metrics.
Research & Coursework
Cardiff University QA Chatbot — Domain RAG System (Coursework Project)
A proposed domain-specific Retrieval-Augmented Generation (RAG) QA system planned as part of MSc coursework, aimed at answering university policy and information queries using document retrieval and LLM generation. The project will evaluate and compare multiple retrieval strategies for accuracy and faithfulness.
Research Papers Studied
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks — Lewis et al.
- Dense Passage Retrieval for Open-Domain Question Answering — Karpukhin et al.
- RAG for Large Language Models: A Survey — Gao et al.
Proposed Study Topic
A Comparative Study of Retrieval Methods for Domain-Specific RAG Question Answering: The Cardiff University Case.