LLM Engineering with LangChain: RAG and Agents

A hands-on course on building production LLM applications. Covers LangChain and LangGraph: chains, prompt templates, output parsers, memory, retrievers; RAG end to end — chunking, embeddings, vector DBs (pgvector, Qdrant, Pinecone), hybrid search, reranking, retrieval quality evaluation; agents and tool calling, planning, error handling and loop limits; structured output and function calling; evaluation (LLM-as-judge, golden datasets, regression tests); cost and latency control (caching, streaming, batching, model selection); guardrails, prompt injection and security; observability (LangSmith, tracing); deployment and monitoring. Practice: build a RAG bot and a tool-using agent.