When Should AI Adoption Be Measured by Business Metrics? — Revisiting AHEAD and LEVER 2/2

Measure harness learning before business value

AI 도입 성과는 언제 비즈니스 지표로 봐야 할까 — AHEAD·LEVER 다시 보기 2/2

Measuring harness learning before business value

AI Wrote the Code Faster—Why Is Review Harder? Reducing Shifted Cognitive Load

Close smaller cycles and review contracts before code

AI로 코드는 빨리 만들었는데 왜 리뷰는 더 힘들까 — 이동한 인지부하 줄이기

Why AI shifts cognitive load from coding to review

Did AI Coding Tools Actually Cut Development Cost? — Getting Started with CTS-SW

Measure AI by end-to-end software delivery cost

AI 코딩 도구가 정말 개발 비용을 줄였을까 — CTS-SW 시작하기

Measuring AI by software delivery cost

Why Detailed PRDs Age So Fast — Separating ALPS and ADR Boundaries

Move implementation authority from PRD to ADR

PRD를 자세히 쓸수록 왜 더 빨리 낡을까 — ALPS와 ADR로 경계 나누기

Moving implementation authority from PRD to ADR

Why Avoiding Failure Comes First

Remove recurring failure modes, then run more learning experiments

망하지 않는 게 먼저인 이유

A personal hypothesis for removing recurring failure factors before using AI to run more learning-focused experiments

How I Built the EncBird Harness Layer by Layer — Harness Engineering in Practice

How harness engineering works in practice, layer by layer

EncBird에 하네스를 한 겹씩 씌워온 과정 — 실전 하네스 엔지니어링

How harness engineering actually works in practice, layer by layer

Why AI Adoption Should Not Start with Token Savings — The 3S Stages 1/2

Match AI investment to each adoption stage

AI 도입은 왜 토큰 절감부터 시작하면 안 될까 — 3S로 단계 나누기 1/2

Matching AI investment to each adoption stage

A Lens for Interpreting Phenomena—and Agentic Engineering

Why HITL removal is my lens for agentic engineering

현상을 해석하는 렌즈, 그리고 에이전틱 엔지니어링

A lens turns variables into constants so a phenomenon becomes predictable, and HITL removal is my lens for agentic engineering

The Next Evolution of Agents Will Come from Smarter Tools

The next evolution of agents will come from smart edges, not smarter centers

에이전트의 다음 진화는 똑똑한 도구에서 온다

The next evolution of agents will come from smart edges, not smarter centers

Agentic Engineering and Transitional Technologies

HITL removal is the direction, and coexistence-oriented tools are transitional

에이전틱 엔지니어링과 과도기적 기술들

HITL removal is the direction, and coexistence-oriented tools are transitional