Support Request: Implementing Custom GPTs for Multiple Organizations with Knowledge Base Integration

ChatGPT Feature requests
,
1

Dear OpenAI Support Team,

I hope this message finds you well. We are currently working on a project that involves leveraging your GPT technology to implement Custom GPTs for multiple organizations within our product. We would appreciate your guidance and support on the following requirements:

Overview:

We are designing a system where each organization using our service will have a separate, smart, custom GPT. These custom GPTs will be tailored to each organization, with the ability to feed their GPT models using a knowledge base in the form of PDFs, DOCX files, and links. Our aim is to ensure that:

  • When a client interacts with the custom GPT, it should first search within the organization’s knowledge base for relevant answers.
  • If the required information is not found, the GPT should then fallback to querying the general ChatGPT database.

Technical Stack:

  • Frontend: Our product’s user interface will be built using React to handle end-user communication with the custom GPTs.
  • Backend: The backend will be developed using Node.js (Express), which will facilitate communication between our UI and the custom GPT models for each organization.
  • Custom GPT Integration: Organizations will have the ability to feed the custom GPT with their knowledge base via our UI, passing this data through our backend Node.js API for real-time updates.

Key Requirements:

  1. Smart GPT for Each Organization: Each organization should have its own GPT, tailored to its specific knowledge base, ensuring customized responses based on organizational content.
  2. Knowledge Base Prioritization: When interacting with clients, the GPT should prioritize retrieving information from the organization’s knowledge base, and only query the general ChatGPT database if needed.
  3. UI Integration: We are building our own React-based UI to handle the chat communication between clients and the custom GPT. This UI will also allow organizations to upload and manage their knowledge base.
  4. Backend Communication: Our Node.js (Express) backend will facilitate interaction between the custom GPT models and the organizational data input, ensuring seamless communication between the front-end UI and the GPTs.
  5. Chat Management: The ability to handle and manage chat conversations between the end client and the custom GPT in real-time, using our custom UI.

We would greatly appreciate any insights, documentation, or recommendations you can provide on how best to implement these custom GPT models while ensuring the smooth integration of the knowledge base functionality and chat handling.

1 Like