search

microsoft / synthetic-rag-index

Service to import data from various sources and index it in AI Search. Increases data relevance and reduces final size by 90%+. Useful for RAG scenarios with LLM. Hosted in Azure with serverless architecture.
Apache License 2.0
24 stars 3 forks source link

Switch to Container Apps requires adaptation #130

Open Laki4Ever79 opened 1 week ago

Laki4Ever79 commented 1 week ago

Hi @clemlesne ,

First of all, thank you and all of the people who contribute here.

Since recent switch over to the Azure Container App implementation over from Function App, it seems there might be some modifications needed to the current implementation of syntax-rag-indexer as it relies on @app.blob_trigger() decorator.

Kind regards, Aleksa

clemlesne commented 1 week ago

Hello, indeed there integrations that requires to be m de in order to keep the same functionalities in Container Apps:

The first two can be achieved with the Azure Storage SDK and the last one with KEDA Scaling Rule.

A straightforward implementation for the first two could be the following:

https://github.com/microsoft/call-center-ai/blob/ff761453362321137a260e0ecce47bbb9558c4b6/app/main.py#L109-L139

Then for the third:

https://github.com/microsoft/call-center-ai/blob/ff761453362321137a260e0ecce47bbb9558c4b6/cicd/bicep/app.bicep#L256

https://github.com/microsoft/call-center-ai/blob/ff761453362321137a260e0ecce47bbb9558c4b6/cicd/bicep/app.bicep#L172-L187

jovankurjak commented 1 week ago

Hi @clemlesne, So with these hints it should be possible to modify current synthetic-rag-index to be compatible again with call-center-ai container app? Or it should be done from scratch ? (just another way how I understood your answer)