Nine projects, one operator: what actually carries over between them

The question I get, in various polite phrasings, is basically: how does one person hold nine of these at once? And the assumption buried inside it is that I've got some reusable core. A shared codebase. A framework. A starter template that makes number nine cheap because numbers one through eight paid for it.

I don't. That's not the thing.

Here's what nine projects actually look like under the hood. tab. and Maison are Expo and React Native. App Marketing OS is Python and FastAPI on SQLite. Confluence has a Python engine with a Next.js front on it. CreatorLens, Fundability, CommunityHQ, ClipForge — Next.js, TypeScript, various databases. Cadence Studios is barely a web app at all; it's a pipeline with a control room bolted to the front.

Almost no code moves between those. I've tried. Every time I've gone to lift something out of one and drop it into another, the shape was wrong by just enough that rewriting it was faster than adapting it. Auth in a native app isn't auth in a web app. The database layer is coupled to how that particular product thinks about its data, which is the whole point of it. A shared utils file across nine projects would be nine sets of compromises pretending to be a convenience.

So what does carry?

The plumbing carries as muscle memory, not as code. I've wired auth nine times. Billing a few times. Deploys, environment variables, the way a database connection string quietly rots when a provider migrates a hostname — I've been through all of it enough that the fifth time takes an hour instead of a day. The artifact doesn't transfer. The knowing-what-you're-doing does. It's the difference between owning a tool and owning the callus.

The verification habit carries, and it's the big one. Checking the work is the real skill, and it transfers completely, because it's not about the domain. Whether the model just wrote a fill-simulation for a trading engine or a review parser for a website generator, the question is identical: what would this look like if it were subtly wrong, and have I actually looked at that? That habit is the single most portable thing I have. It's also the thing that took longest to build and would be the hardest to explain to someone starting today.

A sense of the shape carries. Before I write anything now, I have a rough map of where the hard part is going to be. Not from planning — from having been wrong about it eight times. The hard part is almost never the feature that sounded hard. It's the boundary: the API you don't control, the review process, the data that arrives messier than the docs implied. Nine rounds of that and you start smelling it early.

And some specific instincts carry, which surprised me. The hard human-approval gate on the Cadence outreach side wasn't a Cadence idea — it came from tab., where the whole product is public receipts about other people's money and you learn very fast that automation touching someone else's reputation needs a person in front of it. Same instinct, different noun. That one moved across projects intact and I didn't notice until later.

Then there's the other list. The things that don't carry, and this is the part I'd actually want a solo builder to hear.

Domain knowledge doesn't carry at all. Fragrance, retail lending, Metro 2 reporting rules, community operations, market microstructure, TikTok's tolerance for volume. Nine products means nine domains, and every single one starts at zero. The model is enormously helpful here and also the most dangerous, because it will hand you confident-sounding domain knowledge at exactly the moment you have no ability to check it. This is where being fast is a liability rather than an asset.

And distribution doesn't carry. Not one bit. This is the honest ceiling on the whole approach. tab. going through App Store review taught me an enormous amount about App Store review. It taught me nothing about getting creators to try CreatorLens. There's no audience compounding across a portfolio like this, because the audiences have nothing to do with each other — fragrance collectors and Skool operators and small business owners and traders are not a market, they're nine markets, and each one is a cold start.

Which is the thing, really. The building got so cheap it stopped being what I budget around, and everything I just listed as "carries over" is a building skill. So of course nine is manageable on that axis. All nine of the skills that transfer are on the side of the ledger that already got cheap.

What doesn't transfer is everything on the other side. Attention, distribution, being the person who tells nine different rooms of strangers that a thing exists. I've written about attention being the expensive resource and I'm not going to relitigate it, but this is the same finding arriving from a different direction: what carries over is exactly the stuff that stopped being the bottleneck.

So the answer to how one person holds nine of these is a little deflating. I hold nine of them because the parts that carry over are the parts that got easy, and the parts that don't carry over are the parts I mostly haven't done yet. That's not a productivity system. That's just where the cost went.

This one's auto-drafted from my notes on a schedule. If a number isn't in the notes, it doesn't show up here — I'd rather leave a blank than make something up.