Knowledge Layer

Context Engineering

Teach how information is assembled, prioritized and supplied to models and agents.

Why This Matters

The model only sees what you put in front of it. Context engineering is how policy, retrieved evidence, tool results, and memory become a designed workspace rather than an unbounded prompt.

Architecture Position

Context lives in the Knowledge Layer, sitting between retrieval and the model, and is assembled again at the Runtime Layer for every tool turn. See it on the architecture map.

Start Here

Core Concepts

Fundamentals

  • Context windows
  • Prompt structure
  • Instruction hierarchy

Context Assembly

  • Dynamic context
  • User context
  • Application context
  • Tool context
  • Retrieved context

Techniques

  • Few-shot examples
  • Context prioritization
  • Context compression
  • Summarization
  • Context caching

Agent Context

  • Agent scratchpads
  • Context for tools
  • Shared context
  • Multi-agent context

Security

  • Context isolation
  • Context poisoning
  • Instruction injection

Architecture

  • Context budgets
  • Context pipelines
  • Prompt vs context engineering

Advanced

Advanced material

Deeper guides for this domain are on the way.

Coming Soon

Architecture Patterns

Applied patterns for this domain live in the pattern library.

Browse architecture patterns

Case Studies

Enterprise walkthroughs that apply this topic in a full system.

Browse case studies

Related Topics

Roadmaps

This topic appears in the Enterprise AI Architect path.

Follow the Enterprise AI Architect roadmap
Go deeper

Learn this as part of the Enterprise AI Architecture course.

Need help applying this architecture to a real system?