Creating a virtual assistant with different functions in the same API

API
, , , ,
1

Hello,
I want to design a virtual assistant to use in my project. I researched whether I could fulfill the criteria I designed for this virtual assistant separately, but I don’t know if I can use it in the same application.

  1. Only answer allowed questions (domain centric questions)
  2. Creating and reminding a meeting on the specified day and time
  3. When I say bring my invoices, download my analytical data (user data, for example my analytical data for YouTube) and report it, it can do this,
  4. It should also be able to bring me the queries I search by performing semantic search (user search- movie etc.)

Can I do all of these in the same API with one open ai key? How should I set up the architecture? Which PlugIns should I use? Also, I will use LLM and the data is very large, so what would be the most practical, fastest and least costly method?

I would be very glad if you could help me, thank you.

2

You have two ways you can interact with the AIs of OpenAI:

Use ChatGPT, the website chatbot: ChatGPT Plus can be enhanced with GPTs, that allow document uploading and querying. ChatGPT likes to answer any question under the sun, but it is only for you, not for other users.

Program with the API. This is better for in-domain scenarios like answering for users only about your products and their prices. The multiple features you discuss are multiple complex layers of programming, along with other techniques such as vector database search on knowledge, that would be needed to supplement the AI’s natural-language answering capabilities.

3

Thank you for answering my question. Yes, I want to use it as an API too, but I want to know, is it possible to design and use all these scenarios in the same API? Will there be a change in cost and performance?