zach.dev
StartupCities
StoresData
Automation.co
LearnClaude.ai
Who?I'm a software engineer and entrepreneur focused on modern web technologies and AI.
Here's an ongoing autobiography, which also shares the story of my by-the-bootstraps "unschooling" education: now the subject of a chapter on grit and resilience in the bestselling book Mindshift by Barbara Oakley.
An angel investor once described my core soft skill in the role of founder or early team member as: "The ability to perceive exactly what needs to be done. And then to do it."
My experience working in difficult environments around the world means that I can be trusted to get things done, even when things go wrong.
In the past, I coined the term "Startup Cities" as co-founder of StartupCities.org and a startup spinoff, both of which focused on why startups should build cities. I now write about Startup Cities at StartupCities.com
I've won several awards for economic research and have been published or interviewed in Virgin Entrepreneur, a16z's Future.com, The Atlantic's CityLab, Foreign Policy, and in academic volumes by Routledge and Palgrave MacMillan.
Wait... what is this site?
This is my personal portfolio, inspired by the question: "What would the opposite of the two-color template developer blog look like?"
Have fun exploring!
Click the Start Menu to learn more.
Contact:hello @ zach.dev
I think there's a new way to use AI for research.
Most people use AI for research in two ways:
- Open up Claude Desktop and say "research X"
- Use a Deep Research API like Gemini's
Deep Research cites its work and goes, well, deep. But it writes papers that read like bad grad school essays. It buries the relevant data in a thousand words.
"Just ask Claude" is flexible and more concise. But you end up with a bunch of random data points from dubious sources.
Most people don't realize that many of Claude's tools (like it's built-in webscraper) are lossy. It summarizes pages with cheap models. So there's lots of hallucination risk. It's slow and there's no data trail.
For StoresData, I've been working on an alternative that I call "Structured Research."
The goal is a fully auditable trail for every fact, flexibility, no hallucinations, and structured data that folds into the concepts of the domain, in my case, consumer brands.
This problem is harder than it sounds.
You end up mimicking how a smart human researcher works. How do you know which website to trust? How do you know that you should verify a particular fact with another data source? Is this fact still true or was it true only in the past? Data is scattered. It's difficult to parse.
And if you want to research a million consumer brands, as I do, you also have to make it fast and cheap.
I'm super excited to share a brand new StoresData powered by this Structured Research approach in the new few weeks.
