Building a Domain Specific GenAI Chatbot with Serverless
summary
Claims that serverless + RAG makes domain-specific GenAI practical and demonstrates it end-to-end: uses Amazon Bedrock (converse API) for LLM calls, Amazon Kendra for embeddings and retrieval, DynamoDB for conversation history, Step Functions to orchestrate parallel Kendra/DynamoDB queries and a Lambda that streams chunks back over an API Gateway WebSocket; deployment via SAM. Demo builds a working chatbot that indexes serverlessland, answers EDA questions and generates SAM/CDK snippets, and supports up to a two‑hour websocket session.
AI-written summary · about
chapters (7)
related talks
32mBuilding a Serverless GenAI Platform
36mWhatsApp Chatbot Serverless on AWS with Generative AI
39mGenerative AI Solution with AWS Serverless and Bedrock
41mServerless LLM with AWS Lambda and LangChain: A Revolution in Application Development
15mServerless & Agentic AI: Better Together
38mCraft your own Generative AI Chatbot with Amazon Bedrock
50mBuild, Deploy, Debug and Tune Serverless Microservices
47mOutsmarting the Genie: Smart Architecture Decisions for GenAI Apps