AI Development

Covers AI/LLM design philosophy, prompt engineering, model operations, and engineering perspectives in the AI era.

13 articles

$When \"Absolute\" Becomes a Trap — Adding an Untestable-Exit to the Testing Phase$

>AI Development_2026.05.04

$When \"Absolute\" Becomes a Trap — Adding an Untestable-Exit to the Testing Phase

//Our AI workflow agent's testing phase kept getting stuck in a 3-strike retry loop whenever acceptance criteria couldn't be verified due to physical constraints. A record of adding an \"untestable exit\" through prompt-only changes, and the lessons it taught us about quality-gate design.

tags:[#AI, #Best Practices, #Automation]

>AI Development_2026.04.22

$Seven Principles for Designing LLM-Embedded Tools: An 11-Day Decision Log from the CARE Project

tags:[#AI, #Best Practices]

>AI Development_2026.02.09

$Don't Ask Ability in Assessments (Part 2: Advanced with LLM) — Making LLM an Observer, Not an Evaluator

tags:[#AI, #ORIMD]

>AI Development_2026.02.02

$Don't Ask Ability in Assessments (Part 1: Basic Questionnaires) — The Structural Paradox of Questionnaire Design

tags:[#Metacognition, #Self-awareness]

>AI Development_2026.01.29

$Why I Decided Not to Adopt Skills: The Courage Not to Jump

tags:[#ai-development, #technical-decision]

>AI Development_2026.01.20

$Finding the Optimal Verification Cycle in AI-Driven Development: When Should Humans and AI Test?

tags:[#AI, #devloop-runner]

>AI Development_2026.01.07

$Who is Readability For? — Questions Beyond Writing Code with AI

tags:[#AI, #Refactoring]

$When \"Absolute\" Becomes a Trap — Adding an Untestable-Exit to the Testing Phase

$Seven Principles for Designing LLM-Embedded Tools: An 11-Day Decision Log from the CARE Project

$Claude Model Migration: Alias Design Proves Its Worth

$Team Leader's Real Job Was Observation — Lessons from the Unresponsive Agent

$Design for LLM Model Transitions: Migration from Codex 5.1 to 5.2

$Reviewing 22 Technical Articles with Claude Code Agent Teams

$Why Claude Code Agent Teams Should Be Understood Through Organizational Theory

$Offshore vs AI-Driven Development: 6 Patterns for 2026

$Don't Ask Ability in Assessments (Part 2: Advanced with LLM) — Making LLM an Observer, Not an Evaluator

$Don't Ask Ability in Assessments (Part 1: Basic Questionnaires) — The Structural Paradox of Questionnaire Design

$Why I Decided Not to Adopt Skills: The Courage Not to Jump

$Finding the Optimal Verification Cycle in AI-Driven Development: When Should Humans and AI Test?

$Who is Readability For? — Questions Beyond Writing Code with AI