In this Post I demonstrate a RAG approach using Amazon Lex and DynamoDB and Hugging Face llm. This approach enables the use of a Lex bot and LLM to answer questions based on knowledge context from a document. This approach uses LlamaIndex for the whole RAG pipeline.
I have uploaded the whole code for the solution to Gumroad if anyone is interested to follow this.
🗒 Full code
https://achaklan.gumroad.com/l/serverlessrag
📃 Demo of the Doc ingestion
https://youtu.be/G79vO7Lavww
🤖 Demo of the Lex Chatbot
https://youtu.be/Q7TtIM1jRmM
Top comments (0)