Multilingual Semantic Search with Reranking on a prepared large vectorized dataset comprising 10 million Wikipedia documents. It supports dense retrieval, keyword search, and hybrid search.

A RAG app with streamlit as UI app, flask as backend api. Bot trả lời về document, về data cụ thể. Bot trả lời về document của công ty, trả lời về tờ hướng dẫn sử dụng hay gì đó. Data: bot có khả năng query để lấy dữ liệu (dạng dữ liệu có cấu trúc) như csv

🩺 Agentic RAG chatbot for medical Q&A on Diabetes, Hypertension & Asthma. Powered by a LangGraph multi-step reasoning agent, hybrid retrieval (dense + BM25 + Cohere rerank), Groq LLaMA 3.3 70B, Qdrant, and Supabase. FastAPI backend on Cloud Run.

ChromaDB-powered RAG server for Signal.

self rag sample

Production RAG system — hybrid search, Cohere reranking, RAGAS regression gate, Langfuse tracing, and confidence guardrails. Eval-gated deployment on AWS Lambda.

Production RAG system for RUSH University System for Health policy retrieval. FastAPI + Next.js + Azure OpenAI + Cohere Rerank.

pdf embedding search sample

Semantic search and Q&A engine for codebases. Ingests GitHub repositories, parses AST function blocks, runs PII redaction, and queries them using RAG (Gemini + Cohere Reranker + pgvector).