Standard multimodal RAG retrieves the wrong images because captions lack document context. Two fixes: context-aware summaries at ingestion, and using the generated answer to select images.

Agentic Context Engineering (ACE) is the practice of deciding what an agent sees at each reasoning step. Covers the context loop, structuring and compaction techniques, cross-session memory, and how to evaluate context quality.

A working reference for the three decisions behind any production agent: the reasoning pattern, the framework, and the memory system.

n8n is a visual, node-based automation platform whose AI Agent node lets you embed LLM-driven reasoning steps inside otherwise deterministic multi-step workflows.

The Microsoft Agent Framework is an open-source SDK and runtime for building and orchestrating production-grade AI agents and multi-agent systems.