Section guide
Context, Memory, and Knowledge
Context assembly, working sets, durable memory, semantic recall, RAG, and knowledge-bound behavior.
Use when the system must assemble context, retrieve evidence, remember facts, or verify knowledge.
Leave with a clear boundary for context, memory, retrieval, citations, and source trust.
A context and memory policy covering sources, freshness, retrieval, storage, and correction.
Reading order
7 chapters- 01 Context Budgets and Working Sets Guide · Intermediate 10-20 min read Builder · Architect
Context Budgets and Working Sets explains how to select, label, compress, retrieve, and drop context before it becomes hidden risk.
- 02 Context Engineering Pattern · Intermediate 12-25 min pattern review Builder · Architect
Context engineering controls what the model sees: instructions, state, retrieval results, tool documentation, memory, examples, and prior messages.
- 03 Memory-Augmented Agent Pattern · Intermediate 12-25 min pattern review Builder · Architect
Memory gives an agent continuity, but it also creates a durable trust boundary.
- 04 Long-Term Episodic Memory Pattern · Intermediate 12-25 min pattern review Builder · Architect
Long-term episodic memory stores events: what happened, when, who was involved, and why it mattered.
- 05 Semantic Recall and RAG Pattern · Intermediate 12-25 min pattern review Builder · Architect
Semantic recall retrieves relevant material by meaning rather than exact keywords. RAG injects retrieved material into context before generation.
- 06 Working Memory Pattern · Intermediate 12-25 min pattern review Builder · Architect
Working memory is compact, typed task state the agent can update and consult during a run.
- 07 Knowledge-Bound Agents Pattern · Intermediate 12-25 min pattern review Builder · Architect
Knowledge-bound agents ground answers and actions in approved sources, policies, and citation rules.