Prototyping Generative AI Use Cases with Amazon Bedrock and Langchain

Open-source, full-stack Retrieval-Augmented Generation (RAG) platform for private document search, contextual chat, and intelligent automation — built with Amazon Bedrock, Weaviate, AWS Lambda, and a future-ready React frontend.

No description available

Complete RAG (Retrieval Augmented Generation) system with AWS Bedrock, DynamoDB, Node.js backend, and React frontend. Processes PDF documents and provides intelligent Q&A with source citations.

Full-stack RAG AI app, built with React.js for the frontend, and Python, Langchain, AWS Bedrock, ChromaDB, and FastAPI for the backend. I used Docker to containerize the frontend and backend, AWS ECR to store the Docker images, and AWS EC2 to run the containers, with Nginx configured on EC2 to manage traffic and SSL/TLS certificates for security.

Privacy-focused photo manager with AI-powered organization and natural language search. Raspberry Pi (Node.js/Express) backend, React frontend via Cloudflare Tunnel. Uses Azure Computer Vision & AWS Rekognition for tagging and face detection, plus a RAG chatbot (Bedrock Claude + Supabase pgvector) for search.

No description available

No description available

No description available

Complete RAG (Retrieval Augmented Generation) solution with AWS Kendra, Bedrock Claude 4, and React frontend for natural language document querying

RAG-powered adaptive quiz platform for AWS AI Practitioner (AIF-C01) certification prep, built with Bedrock, Pinecone, FastAPI, and React

AI-powered document chat using RAG. Upload PDFs and ask questions about them. Built with AWS Bedrock (Claude 4 + Titan), RDS PostgreSQL + pgvector, React, and Terraform.

Serverless Vectorless RAG on AWS — upload documents, ask questions, get grounded answers using LLM reasoning instead of embeddings or vector databases. Built with Amazon Bedrock (Claude 3 Haiku), Lambda, DynamoDB, API Gateway, React, and Terraform.

This project is a Retrieval-Augmented Generation (RAG) system utilizing AWS Bedrock and FAISS for efficient document retrieval and question-answering. The application consists of a backend built with FastAPI and a frontend built with React. If you try to open the website it will not show any context because I have not uploaded any backend services.

It is a full-stack AI customer support chatbot built using RAG architecture. Answers questions from a private knowledge base using vector search. Built with Python Flask, React.js, ChromaDB, WebSockets, and Groq (Llama 3). Features real-time streaming, source attribution, and a clean chat UI. Same architecture as AWS Bedrock Knowledge Bases.