Distributed C++20 microservices architecture for real-time detection and correlation of DDoS and ransomware activity. Deterministic ingestion, idempotent replay, FAISS-based semantic indexing, and zero-coordination incident correlation via derived trace identifiers.

RAG-based conversational AI for Superior University — 4,750+ pages indexed using LangChain, FAISS, and Groq LLM

This repository contains my internship project work at Flexbox Technologies. I have developed a system that fills the patient details form automatically with the patient data extracted from pdf file.

DocuQuery is a document querying application that utilizes Retrieval-Augmented Generation (RAG) and the Llama2 model for efficient information retrieval from large documents. This user-friendly tool supports PDF uploads and delivers quick, accurate responses to user queries.

SpeciFic is an NLP fanfiction recommender for AO3 that compares Knowledge Graph and semantic-embedding (FAISS) retrieval approaches in an automated evaluation framework.

MedPrompt AI Chatbot is an advanced medical assistant designed to provide information based on medical contexts. It utilizes a Retrieval-Augmented Generation (RAG) approach, leveraging Google's Gemini and Mistral AI models, alongside a FAISS-indexed medical knowledge base for accurate and relevant responses.

Local ChatGPT using Ollama + Python (Streaming + PDF RAG + Embeddings + FAISS + Chat UI)

Self-hosted AI assistant with persistent semantic memory. Remembers past conversations using FAISS vector search and nomic embeddings. Runs 100% locally via Ollama. No cloud, no telemetry.

A production-grade Retrieval-Augmented Generation system delivering accurate, context-aware responses for specialized domains using FAISS vector search, Google Gemini 2.5 Flash, and persistent MongoDB sessions.

AI-powered tutoring system for Indian state-board students. Upload textbook PDFs and get curriculum-aligned answers instantly. Uses Context Pruning to score and filter chapters before querying Gemini LLM — reducing API costs by ~80%. Built with Python, Flask, FAISS, sentence-transformers, and Gemini 2.5 Flash.

This project allows users to upload PDF files and ask questions about their contents using a chat interface with Gemini API.