Why Saturday Robot Exists

This February, I spent about $5,000 on AI tokens.¹

It sounds like a lot, but as I went through what I spent the tokens on, I realized my small team had delivered what would have taken a team 3x the size a year or two ago. Most of the month ran through Claude Code and Codex. People on the team were doing work outside their usual roles.

Count the products launched for sales and marketing teams this year alone; every one is a new tool and a new way of working you now have to learn.

Software teams already kept records AI could use

Almost eight years ago I was head of growth at Kite, an AI coding assistant that existed years before ChatGPT.² The models then could autocomplete and hint, and that was it. Nothing could reason with you yet. We charged for tokens before most people knew what a token was.

I later realized why AI tools were adopted so quickly in software: engineers already kept their work in text. For decades, the code had been the single source of truth, the docs got maintained, the pull requests got argued in public, and mistakes left a record. When models learned to reason, they could use those records to help with the work.

The marketing context was harder for a model to use. Strategy lived in an outdated deck, positioning was a paragraph somebody half-remembered from an offsite, and the playbooks were PDFs nobody opened. Little of that context was readily available to a model.

The same shift is now happening in other functions

Last year I built a go-to-market team where nobody had written a SQL query on day one. By the end of the year we were opening PRs in GitHub every week.

There was no elaborate workflow behind it. As we used the tools, we improved the context we gave them and learned how to use them better.

When Block cut roughly 4,000 people this February, Jack Dorsey wrote that intelligence tools paired with smaller, flatter teams were changing what it means to run a company.⁴ Not every layoff is caused by AI. But the market has an incentive to tell that story, and companies may cut staff based on what they expect AI to do, before it can do the work.

Why write about this?

The best thinking I've read on any of this comes from practitioners writing as they go: Paul Graham on writing as thinking,⁵ or Simon Willison, who ships a tool and has the blog post up the same week.⁶ Ben Thompson explaining the business changes behind an earnings report.⁷ Almost everything else is a 101 explainer or a 2030 prediction. I find less writing that helps me decide what to try at work.

I also ran monthly workshops on AI in content marketing this past year, with thousands of registrants across them. The people who learned fastest all used some form of shared system. The people still stuck were running the same prompt-per-task loop they'd been running six months earlier, and nobody had shown them what a working setup looks like.

What I'll write about

Every post will be about the work itself: how a thing got done, what made it hard, what broke, what I'm still figuring out. Mostly field reports from my own work, sometimes someone whose setup taught me something. If I don't have anything substantial, the piece will be short. When I do, I'll go deep. Each post should give you something specific to try at work that week.

That includes this post. Shipping with it is a starter kit for a knowledge base you can run agents on:⁸ templates and examples you copy over and have running in an afternoon. By Monday you can have your first agent reading your own work.

The name refers to building AI projects on weekends.

A note: the starter kit goes live Monday, May 11. Parts of the site are still rough, I'm fixing things as I go.