Notes on building the AI-native web.
Page 12 of 17
What an AI employee can and can’t do in its first month
Hire it like a junior with perfect recall and no judgment. That framing predicts almost everything about how the first month goes.
Should this run in real time or overnight?
Real-time is the default assumption and often the expensive one. Much AI work is genuinely happier in a batch at 2am.
Your prompts belong in the repository
A prompt is the specification of how your feature behaves. Keeping it in a dashboard where anyone can edit it untracked is a strange choice.
The admin user problem
Every system starts with two roles: normal and everything. The gap between them is where most access problems live.
How to brief a writer on AIEO
Hand them a keyword and you get keyword content. The brief that produces citable writing looks quite different.
The integration nobody owns
Every company has one: the script connecting two systems that one person wrote and nobody has looked at since. Here’s what it costs you.
Do you need a content calendar?
You need a queue and a review cycle. A calendar that mandates publishing on a schedule produces filler, which is now actively harmful.
Scheduled jobs: the code nobody watches
Nightly tasks fail silently for months because nothing complains. The fix is alerting on absence, not on errors.
How to structure a services page buyers can compare
Buyers are comparing you whether you help them or not. A page that makes it easy wins more than one that makes it flattering.
AI employees in a regulated business
The constraints are real and mostly about evidence rather than prohibition. What you have to be able to show, and what it changes.
Questions to ask a software vendor before you buy
The demo answers none of them. Eight questions whose answers determine what the next three years cost.
The three places AI belongs in a support workflow
Everyone puts the bot at the front door. The higher-return positions are further back, where customers never see it.