Azure-Samples/miyagi

# Technical Summary for miyagi Demonstrates design patterns for enterprise AI applications including retrieval-augmented generation, chain-of-thought prompting, vector-based memory with pgvector and Azure AI Search, and fine-tuned foundation models (Llama2, Phi-2). Built on event-driven microservices architecture using Semantic Kernel, Promptflow, LangChain, and LlamaIndex for LLM orchestration, with use cases spanning personalized financial coaching, summarization, and agent-based workflows. Includes implementations across multiple frameworks—Assistants API, AutoGen agents, VSCode extension plugins—deployed on Azure Container Apps and AKS for cloud-native scalability.

752 stars. No commits in the last 6 months.