Posts
All the articles I've posted.
-
The Day Before Launch, I Overhauled Next.js Performance — TTFB, LCP, and One Thing I Never Fixed
A tuning retrospective from the day before launching a solo-developed app: TTFB 3.15s → 10ms, LCP 22.1s → 1.4s. And the story of the PWA bottom system bar color I burned a whole day on but never fixed on the web.
-
The Screen Froze for a Moment When I Tapped — Tracing a Next.js INP Issue
Tracing an intermittent tab-switch stutter, I confused my tools (Lighthouse ≠ INP), misdiagnosed Realtime, fixed an obvious waste with little payoff, and concluded it wasn't a single bug but a structural cost of SPA navigation. A retrospective that keeps the failures in.
-
A Reusable Kit for Solo Development — CLAUDE.md Constitution File + 2-Agent Workflow
Solo development has no reviewer, PM, or QA. This setup encodes those roles into (1) a CLAUDE.md constitution file and (2) two sub-agents — code-writer and code-reviewer — so the machine enforces discipline. Pulled from a recent project's refined setup, keeping only the parts portable to other projects — a CLAUDE.md skeleton, a 6-step 2-agent workflow with 3 operating modes (review-only / finalize / review-and-finalize), agent prompt templates, and a project-agnostic checklist of working principles. Permission scope is deliberately split (writer has no git, reviewer has no Write) so every commit always passes through a review gate. You can drop this into a new project in 5 minutes.
-
Spring Boot #2 — CRUD with REST API · Layered Architecture · Global Exception Handler
Following Spring Boot #1 (first run · DI), #2 covers CRUD with REST API. Starting from the definition that REST is a combination of URL + HTTP method (GET/POST/PUT/DELETE), I completed todos CRUD by attaching @GetMapping · @PostMapping · @PutMapping · @DeleteMapping to the controller. I organized curl options (-X method · -H format · -d data) and verified the flow with a 5-step curl sequence (create → list all → update → delete → final check). Since the controller was holding all the logic, I split it into a service → layered architecture (Controller = request/response / Service = logic / Repository = storage). For exception handling: custom exception → 404 response, and with a Global Exception Handler in place, the controller just throws while the handler takes care of the response mapping — extending into practice the "throw forces propagation" contract I learned in the Java #3 post. Two trial-and-error incidents (trying to start the server without a referenced class present · writing package code before adding the library) got folded into my study guidelines to prevent recurrence.