nzuguem / nzuguem-github-bot

LLM-based Github App using Quarkus Github App and Quarkus Langchain4j extensions
MIT License
5 stars 0 forks source link

Dev UI support #22

Open nzuguem opened 8 months ago

nzuguem commented 8 months ago

For starters, I plan to have:

View information about declarative AI services and tools Add embeddings into the store, if it exists in the CDI container Search for relevant embeddings and view them as a table

nzuguem-github-app-dev[bot] commented 8 months ago

Hello,

Thank you for raising this issue regarding the development of a Dev UI to support declarative AI services and tools. Based on the features you've outlined, it seems you are looking to create a user interface that will allow developers to interact with AI services more efficiently during the development process.

Here's a potential approach to address your requirements:

  1. View Information About Declarative AI Services and Tools: To view information about the AI services and tools, the Dev UI could provide a dashboard that lists all the available services. This dashboard could display key metadata such as service names, types, descriptions, and statuses. You can leverage the Quarkus Dev UI extension mechanism to create a custom page that integrates with your AI services.

  2. Add Embeddings into the Store: For adding embeddings into the store, you can provide a form or an API endpoint within the Dev UI that allows users to upload or specify the embeddings they want to store. If the embeddings exist in the CDI (Contexts and Dependency Injection) container, you can use CDI events or programmatic lookup to retrieve and store them appropriately.

  3. Search for Relevant Embeddings and View Them as a Table: Implementing a search feature within the Dev UI would require a backend service capable of indexing and querying embeddings. You can create a search interface that allows users to input search criteria. The results can then be displayed in a tabular format, with options to sort and filter based on different attributes of the embeddings.

Regarding the practicality of the existing Replay UI, it's great to hear that Quarkus offers an integrated Replay UI when developing in dev mode. This tool can be very useful for inspecting payloads and replaying events. However, if you find that it lacks certain features such as search functionality or immediate event detail availability, you might consider extending the existing Replay UI or building additional tools that complement it.

For example, you could:

To access the Quarkus integrated Replay UI, you can navigate to http://localhost:8080/replay/ while your application is running in dev mode. This should give you a starting point to inspect and replay events.

Please let me know if you need further assistance or if there are specific technical challenges you are facing in implementing these features. Our community is here to help you make the Dev UI support a robust and developer-friendly tool.

Best regards, [Your Name]

Source : https://docs.quarkiverse.io/quarkus-github-app/dev/