LLM App Development Using PromptFlow and Azure Cosmos DB
summary
Shows a concrete integration of PromptFlow with Azure Cosmos DB MongoDB vCore and Postgres using the lightweight pf-azuredata package, with pip install and open‑source repo links. Builds a RAG pipeline over financial transcripts in Azure Blob, preprocesses and stores embeddings in a table named msfd_transcript, and uses Ada‑002 embeddings plus Azure OpenAI for LLM calls. Demo retrieves three results, fixes a wrong answer by changing top K=3, and includes batch evaluation notebooks and end‑to‑end PromptFlow orchestration.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:26Agenda
- 1:34PromptFlow
- 3:33Vector Search
- 5:03PromptFlow Setup
- 6:23Project Overview
- 8:25Establish Connections
- 11:25Preprocessing
- 13:47Retrieval
- 17:22Orchestration
- 23:03Takeaways
related talks
42mCode First LLMOps With Azure AI & Prompty
43mBuild and Maintain your Company Copilot with Azure ML and GPT-4 Q&A
01h 58mBuild, Evaluate and Deploy a RAG-Based Retail Copilot with Azure AI
46mBuild and maintain your company Copilot with Azure ML and GPT-4
36mLLMOps in Practice- Building Secure, Governed Pipelines 4 LLMs
27mBuild a RAG-based Retail Copilot Code-First on Azure AI Foundry
48mTake an Azure OpenAI chat application from PoC to enterprise-ready
29mYour Prompt Is Not a Unit Test — instrumenting LLM apps in production