The easiest way to read the tiny-team story is that AI lets companies shrink headcount.

That is partly true. Base44, the AI app builder Wix acquired for $80 million, is a real signal. Small teams can now ship things that used to require much larger teams.

But the more useful question is not only what work disappears.

It is what responsibility moves somewhere harder to see.

In this episode of Zero Drag, JJ looks at Base44, solo-founder pressure, and Y Combinator’s advice on making a company legible to AI. The episode is about the part of automation most businesses skip: before you remove the person, know what weird stuff they were catching.

The routine work may shrink. The first draft, the support answer, the report, the handoff, the basic task can all get lighter.

But the customer issue that does not fit the script still needs an owner.

The security scare still needs judgment.

The complaint that looks small, but points to something bigger, still needs someone who knows what matters and what would be expensive to get wrong.

That is the trap inside the tiny-team dream.

AI can handle more of the work. Someone still has to own the weird stuff.

In This Episode

* Why the Base44 story is not only a headcount story.

* What changes when AI clears routine work but edge cases remain.

* Why “making the company legible to AI” is really an operating design problem.

* How to inspect a role before assuming AI can safely shrink it.

Referenced

* Fortune: https://fortune.com/2026/05/18/solo-founders-ai-automation-entire-teams-entrepreneurs/

* Lenny’s Podcast / Maor Shlomo:

* Axios: https://www.axios.com/2026/06/09/ai-entrepreneurs-founders-nasdaq

* Nasdaq Economic Institute: https://www.nasdaq.com/economic-institute/ai-enabling-more-entrepreneurship

* Y Combinator:

Try This

Pick one role, workflow, or customer moment you think AI can shrink. Pull the last ten times the normal path broke and someone had to step in.

That pile shows what the system needs to learn before you remove the person.

Most AI demos show the easy version: a lead comes in, the system responds, the task moves forward. This episode is about what happens after that, when the customer needs something weird, the form answer technically fits but makes no sense, or the automation runs into the edge of policy.

In this episode:

• Why the happy path is not the real test of an AI workflow.

• What business owners are really asking for when they say they want AI to just work.

• Why one reliable agent pointed at a real workflow can beat a pile of agents someone has to monitor.

• How the shift from departments to revenue loops raises a harder question: where does the business store what it learns?

• A practical diagnostic: pull the last ten exceptions from a workflow that feels mostly automated and see where the judgment actually lives.

Referenced:

• Greg Isenberg / Nick from Orgo clip: https://x.com/starmexxx/status/2056686555083706690

• Eric Siu AI-forward agency org-chart thread: https://x.com/ericosiu/status/2058227707859935536

Try this:

Pick one workflow that feels mostly automated. Pull the last ten exceptions and ask: who caught them, what did they know, and did the business learn from it?