semantic grep
A professional-grade Retrieval-Augmented Generation (RAG) platform that transforms your documents into an interactive knowledge base. Built for high-performance semantic search and context-aware conversations.
Lightweight cross-encoder reranker for e-commerce search, fine-tuned on Amazon ESCI. Same size as the baseline, measurably better.
MCP server exposing hybrid RAG (BM25 + dense + RRF + reranking) over SEC filings — with FastAPI backend, Next.js frontend, and 22-query evaluation.
AI candidate ranking system — hybrid retrieval, cross-encoder reranking, and multi-signal scoring for the Redrob Data & AI Challenge 2026
full-stack legal document RAG platform using FastAPI, React, Gemini, and Qdrant.
Retrieval-Augmented Academic Integrity Platform featuring multi-source plagiarism detection, semantic retrieval, AI writing pattern analysis, originality enhancement, and explainable AI reports.
Hybrid-search RAG assistant with independently verified citations, built from scratch (no LangChain) over 241 arXiv papers on LLMs, RAG, and agentic AI.
Most AI chatbots just guess answers based on a simple search. Enterprise RAG Orchestrator is built differently. It is an advanced and a highly secure AI-assistant for engineers. It combines document search with live database querying, protected by a 9-layer security wall and self-correcting logic.
Multi-stage retrieval over ~90k arXiv papers — BM25L, embedding search, and cross-encoder reranking
Mourafiq is a multilingual AI assistant designed to help Moroccan citizens understand and prepare their administrative procedures. The system uses a Retrieval-Augmented Generation (RAG) architecture to retrieve relevant administrative information and generate structured answers in the user's language.
A production-ready Medical AI Chatbot powered by RAG (Retrieval-Augmented Generation). Features hybrid BM25/MMR retrieval, multi-turn conversation memory, and clinical safety filters. Built with Flask, LangChain, Groq (Llama 3.3), and Pinecone.
Turn any website into a conversational AI assistant, auto-crawls a site, builds a RAG pipeline with local embeddings + reranking, and answers questions grounded strictly in the site's real content.
As RAG is an important portion of my fyp project named as DBRAIN, so I make my hands dirty around these technologies in depth by making an AI co-pilot that answers database questions using official docs. Includes strict quality checks, accurate citations, and a zero-hallucination policy.
Upload a PDF, ask questions about it, and get answers that cite the pages they came from.
Production-Grade Enterprise Knowledge Base Assistant** designed to help users quickly search, retrieve, and understand enterprise documentation using **Advanced Retrieval-Augmented Generation (RAG)** architecture.
Local-Hybrid RAG pipeline with Next.js frontend, stringent RBAC, Qdrant hybrid search, local cross-encoder reranking and automated citation verification.
Resume Shortlister with custom Job Description and Candidates.json profiles
A distributed, enterprise-grade Hybrid RAG platform that solves semantic search blind spots and ingestion bottlenecks through asynchronous task processing and two-stage information retrieval.
HireSense.ai is an AI-driven talent intelligence platform that leverages semantic embeddings, RAG pipelines, and LLM-based re-ranking to evaluate candidate–job compatibility beyond traditional keyword matching.
To associate your repository with the cross-encoder-re-ranking topic, visit your repo's landing page and select "manage topics."