How to Create a Generative Chat App using your Documents

Published 2024-03-19
Learn how to easily create Generative AI chatbot for your website using your own documents using Google Cloud's Vertex AI Conversation. In this video, you will learn how to use your Documents like pdf,html, unstructured data etc to power your Generative AI Chatbot.

#GenAI #GoogleCloud #VertexAI

All Comments (14)
  • @asharkhalid6934
    Hey. The video was great. It really help me to understand how gen AI works with document. I have use case I want to manage an e-commerce site with the gen AI chatbot. I want to use my document as a product. Could you share some tips how can I manage them in more efficient way ?? Thank you in advance.
  • @ianstack8013
    These are great videos thank you so much! I wanted to ask if you think you will make a video on how to incorporate more than 1 data store with an agent? Such as documents from the cloud storage and an indexed website?
  • @rajivraghu9857
    Excellent tutorial sach! As part of the dialogflow cx, can i integrate custom REST API endpoints(Function calling) Ex: I have a usecase in my application to find the order details based on Order Number..
  • hey, firstly an excellent video with detailed steps. Really useful. I was able to follow it and got the desired result. However my end use case is integrating with slackbot as all my users are in slack. Can you help me understand how to do it
  • Hello, thanks foe sharing this intuitive video, I have couple questions: 1. What about using multiple bigquery tables (for example 3-4 tables that have different information) to build chat app? Also, Can I choose multiple datastores? 2. Do you have a demo recording using bigquery tables that you can share?
  • @SagarAlwani
    Hi, amazing videos, Can you please tell me one thing: I want to limit my bot to my contents only and it should not respond on questions like: "What is the price of Bitcoin?" or "Will it rain in LA?" Can you please help?
  • @ArminAmirbeigi
    Hi , It was a grate video and tutorial. I built my agent to connect to my Cloud storage to access documents to help support team. .It works as expected just a minor bug. I can't click on the link that has been provided by chatbot. It gave me "403 Forbidden" Error. My storage is not public and I am using corporate google console platform. The Source link displayed along with responses doesn't work for me, could you please advise How can I fix this issue
  • Thank you very much. Is it possible to create as mobile app with similar chat bot
  • @ricardosouza6138
    Hi there! Great tutorial. I would like to know if it is possible to hide or not display the "custom payload" return in the chat! (In production) In my case, I would only like the answers, but without having to show the reference of where the bot got the information!
  • @DionaMiranda
    Hi, Thanks you so much for the wonderful video. It helps me a lot. Could you please tell me if I add new files in the GCS bucket(or new entries in the BQ table) for data store, will the agent accept the new data? Will it work? We will be having a new business data.
  • @aryansinha6701
    sir I had a question, how do I fine tune the responses
  • @lier9832
    Hello. I've done all of the steps in your video, but instead of html pages I uploaded 2 pdf books. It was fully indexed but after trying out Dialogflow Messeger or even simply click "Test Agent" Agent does not understand questions about books and sends me back default response (Sorry, as an AI assistant in a company, I can only help you on the company website. How can I help you today?). What have I done wrong or why doesn't it work?