Chat with your own documents and get to-the-point answers

The scenario parses your stored files, tables included, builds a searchable memory and answers chat questions using only that content.

• Answers come only from your documents, nothing invented
• Tables inside PDFs get parsed, not dropped
• New files in AWS S3 are picked up on each run
• Search across the whole pile takes seconds

AWS S3Embeddings OpenAIDefault Data LoaderPinecone Vector Store