Ziyaad
I started building because I wanted things that did not exist yet.
Most of what I know came from making something badly, breaking it, and reading until I understood why. I use AI heavily, but I type the code myself — if I cannot explain a line back, it does not go in.
I care about work that survives being used. A site that only looks good in a screenshot is not finished. A device that works twice and then jams is not a device. That is the bar I am aiming at, and I miss it often.
How I work
- 01
Separate the problems
Nothing should be asked to be two things at once. Give each job to whatever is actually good at it.
- 02
Prove the pieces alone
A build you cannot take apart is a build you cannot debug. Each part earns its place first.
- 03
Accept the cost out loud
Every choice has a price. Say what it is instead of pretending the design has no downside.
Ruled out — do not relitigate
- Code I cannot read backI have to be able to fix it at 2am
- Comparing frameworks firstNothing gets built that way
- Portfolios that describe the workShow it running or it did not happen
- Waiting until it is perfectIt never is, and then it never ships