Automatically update Azure AI search indexes from BLOB storage for use in Azure AI studio - RAG pattern queries

Question

Hello,

I'm currently using Azure AI studio to upload PDF files to a BLOB store, create an index, and then query this data using Chat GPT 3.5 and the RAG pattern without any coding. It is working great!

However, I'm trying to set up the architecture to update the Azure AI Search index automatically whenever the data source changes. The desired behavior is when I add or remove a file from the BLOB storage, the answers from the LLM are up to date based on the new data.

I have tried creating an index and an indexer for my data directly in Azure AI search. In Azure search, I'm doing an "Import and vectorize" action and linking this to the Azure AI studio playground using the Chat GPT3.5 model. But, I get an error when I try to send a prompt in the playground.

(Validation error at #/data_sources/0/AzureCognitiveSearch/parameters/fieldsMapping/contentFields/0: Input should be a valid string

A quick search indicates that this could be a field mapping issue. Any pointers here on how to remedy this would be appreciated.

I've also tried incrementing the index version manually in AI studio, which works. But, if there was some way to automate this process, it would also solve my problem.

End result I'm looking for is to create my own chat client and use the same API (completions) as the playground.

Thanks in advance