Why AI cleanup should not be the extraction engine
How PromptReady changed AI cleanup from raw HTML rewriting into a quality-gated pass over a local Markdown baseline.
Writing archive Page 2
Older notes, retrospectives, and explanations from project work.
Why semantic code search is not enough for coding agents
Notes on building Satori around deterministic MCP workflows, exact reads, stale-index warnings, and bounded retrieval contracts.
When index cleanup lies, agents make unsafe decisions
A practical note on debugging Satori's clear/reindex path by treating remote vector state as an explicit lifecycle contract.
Why MCP CLIs fail when stdout is treated like logs
Notes on turning Satori's shell CLI into a disciplined MCP wrapper with stdout safety, dynamic bootstrap, and deterministic exit behavior.
Why this portfolio is a public engineering notebook
A practical note on making hamza.my.id useful through project evidence, technical notes, and restrained design instead of generic portfolio polish.
Why frontend-heavy engineers need backend contract discipline
Notes on moving deeper into backend work by treating APIs, validation, schema design, and failure states as product contracts.
Why "full-stack" is only useful when the product path is clear
A practical framing of moving from frontend depth into full-stack product engineering without pretending every layer has equal depth.
Why a 1500+ article migration needed a workflow, not just Puppeteer
Notes from turning a large WordPress migration into a resumable Node.js automation pipeline with logs, checkpoints, and browser boundaries.
Why state-heavy interfaces fail without named events
Practical lessons from Angular and NgRx on making product behavior visible through actions, reducers, selectors, and predictable UI states.