Your question is Debug Multimodal Pipeline with CosmosDB Issue. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
We have a multimodal chatbot that processes documents containing both text and images. We use a Vision Model for image summarisation and store everything in Azure CosmosDB (NoSQL). The pipeline works fine initially, but after processing 15-16 chunks, we encounter a 'CosmosDB not found' error, but only for documents containing both text and images. Debug it.
Asked in the Round 4 stage. The interviewer answered clarifying questions and suggested considering the async factor. Reported follow-ups: Why specifically a 'DB not found' error? Are text and image summarisation payloads sent separately or merged?