HW
Hannah Weiss
2 weeks ago
This is the resource I wish I had when I started. Clear mental models, zero fluff.
Free preview available. Sign in and subscribe to unlock the full book.
Vizuara AI Labs · intermediate
AI Context Engineering
Engineer the context window like a system.
Move past prompt engineering to treat everything an LLM sees — instructions, tools, retrieved knowledge, and memory — as one engineered system. Learn the four strategies (write, select, compress, isolate), then build RAG, MCP tools, memory, and agents from scratch.
43 capsules157 figures~8 hoursby Dr. Raj Dandekar
Read on your Kindle or e-reader
Download the EPUB and read offline — perfect for the train. Works on Kindle, Kobo, Apple Books & more.
00Context Foundations6 capsules
Context Foundations — 6 chapters.
01Why Context Engineering, Not Prompt Engineeringconceptfree10 min02Software 1.0, 2.0, and 3.0conceptfree10 min03Anatomy of the Context Windowconcept🔒11 min04The Four Strategies: Write, Select, Compress, Isolateconcept🔒12 min05Context Failure Modes: Poisoning, Distraction, Confusion, Clashdeep-dive🔒14 min06Your First Context Lab: Measuring Tokens and Effectscode🔒9 min01The Instructional Layer5 capsules
The Instructional Layer — 5 chapters.
07System Prompts as Programsconcept🔒11 min08Instructions, Examples, and Guardrailsintuition🔒13 min09Designing Roles and Personasintuition🔒10 min10Structured Output and Formatting Contractscode🔒11 min11Instructional Layer Labcode🔒12 min02Write and Select: RAG7 capsules
Write and Select: RAG — 7 chapters.
12Write vs Select: Two Halves of Knowledge Injectionconcept🔒12 min13Why RAG Exists: Grounding, Freshness, and Costconcept🔒12 min14Chunking and Embeddingsmath🔒12 min15Vector Search and Retrievalcode🔒12 min16Reranking and Hybrid Searchdeep-dive🔒11 min17Building a RAG Pipeline End-to-Endproject🔒15 min18RAG Lab and Interactive Quizcode🔒14 min03Tools and MCP6 capsules
Tools and MCP — 6 chapters.
19Tools as Context: Giving the Model Handsconcept🔒13 min20Tool-Calling Mechanicscode🔒11 min21What Is MCP? The Model Context Protocolconcept🔒10 min22MCP Servers and Clientscode🔒13 min23Building a Useful MCP Server and Clientproject🔒12 min24Tool-Context Hygienedeep-dive🔒12 min04Compress and Isolate5 capsules
Compress and Isolate — 5 chapters.
25Why Compress: The Token Budget Is Finiteconcept🔒11 min26Summarization and Pruning Strategiesintuition🔒12 min27Isolation: Quarantining Context Across Sub-Agentsconcept🔒12 min28Multi-Agent Context Boundariesdeep-dive🔒12 min29Compress and Isolate Labcode🔒12 min05Memory Architectures5 capsules
Memory Architectures — 5 chapters.
30The Stateless LLM Problemconcept🔒11 min31Short-Term vs Long-Term Memoryconcept🔒12 min32Episodic, Semantic, and Procedural Memoryintuition🔒12 min33Building Memory with Retrievalcode🔒10 min34Memory Architectures Labcode🔒14 min06Setting Up the Build Workflow4 capsules
Setting Up the Build Workflow — 4 chapters.
35A Coding-Agent Build Workflowconcept🔒9 min36Claude Code and OpenClawconcept🔒11 min37Context Files and Repo Hygienecode🔒12 min38Workflow Setup Labproject🔒11 min07Capstone Builds5 capsules
Capstone Builds — 5 chapters.
39Capstone: Building an Email Agentproject🔒11 min40Mega Build: Driving Claude Code End-to-Endproject🔒10 min41OpenClaw and RL Agentsdeep-dive🔒11 min42Building Conversational Voice Agents for Bharatproject🔒13 min43Putting It All Together: A Context-Engineering Playbookconcept🔒12 minRatings & reviews
751 readersRate this book
EP
Elena Petrova
4 months ago
Beautifully produced and genuinely deep. The reader experience makes it easy to keep going for hours.
ML
Mei Lin
7 months ago
Loved the from-first-principles approach. It rebuilt my intuition rather than just handing me formulas.