aws-solutions-library-samples/qnabot-on-aws

Guidance for QnABot on AWS is a multi-channel, multi-language conversational interface (chatbot) that responds to your customer's questions, answers, and feedback. The solution allows you to deploy a fully functional chatbot across multiple channels including chat, voice, SMS and Amazon Alexa.

What it solves

QnABot on AWS provides a multi-channel, multi-language conversational interface to automate customer support. It reduces call center wait times and improves customer experience by providing personalized tutorials and intelligent question-and-answer support across chat, voice, SMS, and Amazon Alexa.

How it works

The system uses a Content Designer UI for admins to manage questions and answers, which are stored in Amazon OpenSearch Service. When a user interacts via Amazon Lex, Alexa, or Amazon Connect, the system processes the request through a fulfillment Lambda function. It leverages Amazon Bedrock for LLM-powered text embeddings and generative responses, Amazon Comprehend and Translate for language handling, and Amazon Kendra or Bedrock Knowledge Bases as fallback mechanisms for retrieving information from external data sources. Guardrails are implemented at the pre-processing and post-processing stages to block harmful content and mask sensitive information.

Who it’s for

It is designed for businesses and organizations that need to deploy a scalable, intelligent chatbot for customer support across multiple communication channels.

Highlights

  • Multi-channel deployment: Supports chat, voice, SMS, and Amazon Alexa.
  • RAG capabilities: Integrates with Amazon Bedrock Knowledge Bases and Amazon Kendra for retrieval-augmented generation.
  • LLM Integration: Uses foundational models via Amazon Bedrock for semantic matching, query generation, and response synthesis.
  • Built-in Guardrails: Includes pre- and post-processing filters to ensure safety and prevent PII leaks.
  • Multi-language support: Uses Amazon Translate and Comprehend to handle non-native language requests.

Related

  • Project
    aws-samples/bedrock-chat

    A multilingual generative AI platform powered by Amazon Bedrock that enables the creation, sharing, and management of custom RAG-enabled bots and AI agents.

  • Project
    liangdabiao/langgraph_multi-agent-rag-customer-support

    A multi-agent RAG customer support system built with LangGraph and LangChain that features security guardrails and a human-in-the-loop approval process for sensitive operations.

  • Project
    Azure-Samples/serverless-chat-langchainjs

    A serverless AI chatbot implementation using LangChain.js and Azure, enabling Retrieval-Augmented Generation (RAG) to answer queries based on enterprise documents.

  • Project
    ChatbotXIO/ChatbotX

    An agentic chat marketing platform that provides a self-hosted, omnichannel alternative to tools like ManyChat for automating customer interactions using AI agents.

  • Project
    yincongcyincong/MuseBot

    MuseBot is an open‑source Go bot that connects Telegram, Discord, Slack, Lark, DingTalk, Enterprise WeChat, QQ and WeChat to a wide range of LLM APIs (OpenAI, DeepSeek, Gemini, OpenRouter, etc.). It streams AI replies, supports image/voice input, function calls, RAG, admin UI, metrics and cron jobs, and can be run locally or via Docker.