A live coding walkthrough of a Spring AI workshop using Spring AI 2.0.0-M4 and Embabel 0.3.4, running entirely locally with Ollama and Redis. The session covers four progressive labs: basic Spring AI chat with system prompts, RAG (retrieval-augmented generation) using Redis as a vector store, an agentic code review workflow using Embabel's goal-oriented action planning, and a code assistant agent. The workshop is designed to run on an 8GB laptop using the Qwen 2.5 Coder model (~900MB) via Docker Compose, requiring no cloud API tokens. Key concepts include LLM hallucination, data cutoff dates, document ingestion, vector embeddings, and type-safe agentic pipelines using Java records.
•1h 22m watch time
Sort: