Record
Log
Articles and build notes, in one place.
4 min readArticle
A practical guide: write AGENTS.md and a task brief before handing your project to a coding agent
Coding agents read your instructions before writing a line. A short guide to AGENTS.md, how tools like Codex and Claude Code read it, and a seven-part task brief you can copy.
5 min readArticle
Arabic AI is moving from models to agents: reading HUMAIN's latest announcements
Three announcements in five weeks: a large Arabic model in research preview, Arabic-speaking agents with Napster, and an operations service with EY. What is actually available today, what is still a promise, and what it means for anyone building Arabic products.
4 min readArticle
When the answer becomes the interface: what ChatGPT's Intelligent UI means for product builders
A teardown of the feature OpenAI shipped with GPT-6: diagrams, buttons and calculators generated inside the answer. What was announced, what we don't know yet, and five practical lessons for product builders.
4 min readArticle
Build log: why I deleted the demo version of “Ask me about you” and built a real self-check
The story of my site's digital-footprint page: it began as an educational simulation with a fabricated file, then I replaced it with a real check of your own accounts only, with three honest states: found, not found, and could not verify.
6 min readArticle
Google launches Playground: when a game becomes a new no-code format
Google Labs launched an experimental platform that turns a text description into a game that runs in the browser. What is actually available, what is still a promise, and why it matters to builders and creators in Saudi Arabia and the region, even though it is US-only for now.
4 min readArticle
JEV: an open decision model after closed Jev opened the door
Verified notes on AutoTrust’s JEV-26B-Decide: open-weights decisions, Gemma 4 base, adaptive thinking, and how it relates to TypeSafe’s Jev without inventing “world’s first” claims.