Tips
One problem, one move, one result. The small things that changed how a week runs, written so you can copy the move without reading an essay first.
Tel Aviv
The tool is never the hard part. Getting people to use it is.
I build the automations and ship the products myself now. n8n, agents and skills, sites deployed straight from Claude and Cursor. Then I work on the half that most AI content skips entirely, which use case is worth doing at all, in what order, and how a team actually changes the way it works.
Why this site exists
Almost everyone writing about AI is standing on one side of a gap.
One side
People who can build
They ship impressive demos. They have never had to walk into a company where four teams disagree, the data is a mess, and the person who has to use the thing is already behind on their week. The demo works. Nothing changes.
The other side
People who can advise
They understand organisations and they can talk about strategy for an hour. They have never built the workflow themselves, so their advice stops exactly where the real problems start, and they cannot tell you what is actually possible this quarter.
Where I stand
This year I crossed from the second side to the first. I finished an AI champions programme, and since then I have been building rather than briefing. Automations in n8n. Agents and skills that do work I used to do by hand. Websites developed and deployed by me, including this one. Nothing here is theory I read about, it is what survived contact with my own week.
The interesting part is that the building turned out to be the easy half. The hard half is still people, sequencing, and knowing which problem is worth solving. That is the part I spent years learning, and it is what I write about here.
Three ways in
Short, medium and long.
Different amounts of your attention. All of it filtered through one question, can someone use this tomorrow morning.
About
Roy Lapin
Thirty one, based in Tel Aviv. I spend my days building with AI and my evenings reading about it, and I have a habit of pulling the two usable ideas out of a long podcast and throwing the rest away. Friends and colleagues started asking me for that summary. This site is where it goes now, so it stops disappearing into private messages.
The background is go to market. I spent four and a half years building the partner ecosystem at monday.com, growing it from around four thousand partners to more than fifteen thousand, hiring and leading the team, and sitting with agencies and services teams in every market you can name. That is where I learned how organisations actually decide things, and why most good ideas die in the second meeting rather than the first.
It is background, not the headline. What I do now is build.
Stay in touch
New pieces by email.
One a week at most, no marketing, unsubscribe in one click. That is the whole promise.
The signup opens with the first published piece. Until then, the pages above are the whole site.