Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature Request]: RAG integration #68

Open
AgentGenie opened this issue Nov 22, 2024 · 4 comments
Open

[Feature Request]: RAG integration #68

AgentGenie opened this issue Nov 22, 2024 · 4 comments
Labels
enhancement New feature or request

Comments

@AgentGenie
Copy link
Collaborator

Is your feature request related to a problem? Please describe.

Integrate the latest RAG tech

Describe the solution you'd like

Current integration list

Falkor DB
MS GraphRAG
Current investigation list

RAPTOR https://arxiv.org/html/2401.18059v1

Additional context

No response

@AgentGenie AgentGenie added the enhancement New feature or request label Nov 22, 2024
@qingyun-wu
Copy link
Collaborator

Hi @AgentGenie, thank you for suggesting this feature! We are prioritizing this feature in the next major release. Can you share your Discord ID so that I can find you on Discord? Thank you!

@anilpillai007
Copy link

anilpillai007 commented Dec 14, 2024

RAG is a common use case, and we should make it accessible to everyone—not just engineers.

Users should be able to upload documents (PDFs), use framework like FastAgency, Waldiez and get immediate value without coding. By limiting this, we exclude non-technical users like product managers and customer service teams.

PDF support is essential—let’s provide plug-and-play functionality that addresses the most common use cases seamlessly. It should also support multiple pdfs with a default implementation (graph db, raptor etc.)
E.g.
PDF with tables, page with columnar content,
PDF with images

Later releases should also support PPTs, Word docs, etc.

@AgentGenie
Copy link
Collaborator Author

@anilpillai007 Thank you for the input. We will support most text formats soon.

For non-technical users, could you articulate your use cases? e.g. is it adhoc or a step in a workflow? do you need a UI support etc?

@anilpillai007
Copy link

anilpillai007 commented Dec 15, 2024

@AgentGenie
I agree it is not an easy use-case. Trying to get 100% done is a last mile problem. There are companies trying to do this. I have a discord conversation in our channel. See here
Use-case: This is not for me, but for a non-tech person (like a product person)
The way I look at it is.

  1. End-State: A non-tech person uploads a bunch of pdfs, and is able to ask questions. In V1. Most examples give super basic pdfs which are not practical in real world. At the minimum, it should be able to understand tables.
  2. Put this in a wrapper like FastAgency, Waldiez
  3. Give ample examples where a user can just clone one example and change the pdf [minimal work for end user]
  4. There should be steps where an end-user can get an OPENAI api key.

This may be too much out of scope for this ticket. If so, at least 1, 2 and 3 can be accomplished if we aim non-tech users with a suitable well developed pdf parser.

Suggestion:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants