Growing Startups: A Series
Approaches, failures, and lessons in making a new business work.
Spending Too Much Money on a Coding Agent
On making use of large thinking models.
For a year, I’d been coding almost every day with Cursor and Claude Sonnet. Anthropic’s 3.5 and 3.7 Sonnet each rightly earned their dominant place on the programming model charts: they were the least-bad coding models yet. In the earliest days of LLMs, there was tremendous interest in ever-larger model...
Testing the Untestable
The four phases of automated evals for LLM-powered features.
I gave a talk version of this article at the first Infer meetup earlier this month. Let’s say you want to build an LLM-powered app. With a modern model and common-sense prompting, it’s easy to get a demo going with reasonable results. Of course, before going live, you test various...
Starting Forestwalk
A wild startup appears.
Last month, I started full-time on a new startup. It’s early days, but we’re having a lot of fun. A startup, fundamentally, is a search for a repeatable, scalable business model. You rapidly try things, run experiments, learn, and iterate your theories about how to build a useful product that...
Pushing the Frontier
If – and when – GPT-5 might eat your lunch
Lately I’ve been working with a lot of teams and founders that are building products on top of LLMs. It’s a lot of fun! To be an AI product engineer today is to constantly ask new questions that impact how you build products. Questions like: “Is there a way we...
Feeding the Baby
On accidentally becoming a CEO.
A lot of startups with first-time founders have unclear roles. When I started my first business, Steamclock’s co-founder Nigel was far more experienced, being ten years my senior. I’d assumed he would take a more CEO-like role, but other than that I’d put little thought into our positions. We were...
Sell First, or Build First?
An age-old startup question.
Recently I’ve been talking to potential customers about a product space I’ve been exploring. One thing I’ve wondered as I do this is, “At what point should we ask our first customers to pay?” Writing on how to develop a SaaS product tends to advocate one of two approaches: “Sell,...
Splitting Services and Product
The story of a plan: take a break, and get focused.
Last year, I realized it was time to switch things up. In 2010, my co-founder Nigel and I started Steamclock. The vision was to build products for clients, and use those profits to fund our own product development. Which worked! Mostly. We’ve built a client business that’s been growing and...
How to Not Build a Social Network
I attempt stop you from making a mistake.
I hear that you’d like to build a new social network. Seems like a good idea, right? Today’s social media is a tire fire, the companies that dominate it rake in billions monthly, and you have a novel concept for a social app that might make people feel less blue,...
Chain of Fools
We dive into smart contracts - and bananas.
I’m officially sick of hearing about blockchain. As the recipient of various “idea for an app” emails every week, I don’t have much patience for get-rich-quick schemes. Suure, you’re going to put hyperlocal photo messaging “on the Bitcoin”, good luck with that, don’t talk to me or my son ever...
The Fall and Rise of Podcasting
Podcasting just keeps getting better.
In February 2005, Ev Williams and Noah Glass launched an exciting new podcasting business called Odeo. Podcasting, or “audioblogging”, was booming. The Great Startup Cynicism of the early 2000s was lifting, and podcasting was the Next Big Thing. The New York Times explained: Mr. Williams, who is 32, helped found...