joewitt commented on PR #8590: URL: https://github.com/apache/nifi/pull/8590#issuecomment-2102943881
I'm curious what the role of langchain there is. Does that get used as the official qdrant python api? Also seems like this is specifically bound to openai for embeddings. We likely need another technique whereby embeddings are generated with a controller service that is thus pluggable or is done prior to arriving at this processor. I've not looked at the other components we have for writing to vector stores but it also seems like we should consider accepting record data as the input to these rather than JSONL specifically. Just some thoughts - doesn't have to stall this current component necessarily -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
