Enterprise-grade, latency-aware 6-stage adaptive RAG platform with Apple HIG web UI. Features dual-filter extraction, Qdrant HNSW + BM25 hybrid retrieval, calibrated RRF (k=20), continuous confidence estimation with top-1 consensus guard, blended cross-encoder reranking, and 97.3% abstention accuracy.
Production ML & AI systems
Twelve end-to-end engineering builds — expand any project inline for architecture, evaluation benchmarks, and learnings.
A lightweight, mobile-first personal technology and research monitor aggregating official company blogs, broad news discovery, and research indexes.
Fine-tuned version of Qwen2.5-3B-Instruct optimized for Data Science and ML coding tasks using QLoRA on 16.4K instruct examples with Unsloth. Achieved an 86% win rate over the base model with +109.5% BLEU and +66.9% ROUGE-L score gains.
AI-powered resume intelligence platform and analysis engine. Evaluates resumes with zero fluff, extracts structured factual profiles via Docling and PyMuPDF, identifies ATS vulnerabilities, critiques bullet points, and benchmarks candidates against 12+ tech roles and job descriptions.
Production-oriented Retrieval-Augmented Generation system for research papers and technical docs. Combines hybrid retrieval (Dense + BM25 + RRF), cross-encoder reranking, evidence gating, and claim-level citation verification to minimize hallucination.
AI-powered hackathon judging platform. Upload a CSV of team submissions and a multi-agent pipeline - five parallel judge agents, a bias auditor, and a chief judge - scores every project automatically, generates written feedback, and produces a ranked leaderboard.
Production-grade FastAPI & AI Command Portal for Tata Power DDL that transforms raw smart meter signals into a probability-ranked enforcement hit-list for targeted electricity theft raids.
Centralized multi-page web dashboard showcasing a compilation of comprehensive data analysis reports (India Crime, CPI Inflation, Delhi Rainfall) with interactive Chart.js analytics and role-based controls.
End-to-end automated ML workflow - dataset upload → cleaning → EDA → training → predictions - with exportable PDF reports and a dataset-aware chatbot.
Full-stack app that analyzes sentiment of YouTube video comments using a custom-trained LightGBM model exposed via FastAPI, paired with a Chrome Extension for on-the-fly analysis directly from the browser.
DL-based stock price forecasting system analyzing 5 years of market data using a 2-layer LSTM with technical indicator feature engineering.
ML-based college prediction system analyzing JoSAA counselling data (2021–2025) with XGBoost rank prediction and chance estimation (Safe/Moderate/Risky).