Ai Engineering For Hyper Personalized Assistants

A dependency-ordered practitioner guide to building personalized LLM assistants, from context-window mechanics through prompt assembly, tool use, MCP, memory architectures, and cost-aware routing.

Textbook

Read the full textbook Long-form companion to the chapter guides →

Chapter Guides

  1. 1 The Anatomy of a Hyper-Personalized Assistant
  2. 2 Tokens and the Context Window
  3. 3 Managing Long Histories: Truncation, Summarization, and Compaction
  4. 4 Prompt Assembly and System Prompts
  5. 5 Tool Use and Function Calling
  6. 6 The Agent Loop: Orchestrating Multi-Step Tool Execution
  7. 7 MCP Fundamentals: The Model Context Protocol
  8. 8 MCP Transports and Pluggable Servers
  9. 9 Short-Term Memory: Session State and Conversation History
  10. 10 Long-Term Memory: User Profiles and Personalization
  11. 11 Retrieval, Embeddings, and RAG for Personalization
  12. 12 Model Routing: Cheap Triage vs. Expensive Reasoning
  13. 13 Cost Control: Caching, Rate Limits, and Production Economics
  14. 14 Putting It Together: Architecture, Evaluation, and Next Steps