Back to columns
Column

Agent Development

18 articles grouped by theme. 16 are in English.

Group 01

Intent & Control

  1. Article 1Agent Intent Recognition: From Classifier to Executable Control System
  2. Article 2Thinking About Intent Recognition: Plan + Tool Design
  3. Article 3Intent Recognition: Unbounded to the Model, Bounded to Code

Group 02

Slots, Practice & Exit Validation

  1. Article 4Slot Extraction in Intent Recognition: The Pitfalls of Chinese Time Parsing
  2. Article 5Why We Deleted 530 Lines of Intent Regex: A Calendar Agent's Architecture Rewrite
  3. Article 6How to Stop an LLM Agent from Faking "Created": Exit Validation That Checks Facts, Not Wording
  4. Article 7I Gave Up the Ops Workbench and Generated a Thousand Calendar Images with a Few Sentences

Group 03

Memory, Retrieval & Compaction

  1. Article 8Hybrid Retrieval for Agent Memory: Dense, Lexical, and RRF
  2. Article 9RAG — Retrieval-Augmented Generation
  3. Article 10Designing Layered Conversation Compaction in Anpai

Group 04

Tools, Protocols & Production

  1. Article 11MCP — Model Context Protocol: Build a Custom Server Cursor Can Call
  2. Article 12MCP Dual Role: Client and Server Together on an AI Testing Platform
  3. Article 13A2A — Agent-to-Agent Protocol
  4. Article 14Integrating DeepSeek Tool APIs into Anpai Architecture and Implementation
  5. Article 15Why I Built an AI Gateway: Model Routing, Cost, and Rate Limits for Indie Developers
  6. Article 16What Separates AI Development from Demo to Production

Group 05

Multi-Agent Orchestration

Chinese articles in this group are published; English versions are not available yet.