Ever since I met @sidd in our early days at Stripe, I hoped we’d one day find an idea big enough to start a company together.
That day is finally here.
After building quietly for the past year, I’m excited to officially launch @AccrualHQ, a new foundation for professional accounting 👇
So October 15th, the extended US tax deadline, is just around the corner, and I have some observations which are more about LLM progress than taxes.
Background: many people professionally involved with LLMs estimate 2026-2028 as the year where one can get an LLM to "do taxes."
Having now spent about half my life in each (and loving both), herewith the pros and cons of Europe and the US in everyday life:
Better in Europe
• Bike lanes and bike infrastructure. London, Paris, and Amsterdam are all excellent these days. (As are many other European cities.) Made even better by easy-to-rent e-bikes—now almost always the fastest way to get around.
• The urban walking experience generally. Partly for density reasons, and partly because of...
• Late-night cafe, brasserie culture. Is there an economic reason for this or is it just climate and contingent zoning?
• Architecture. Around 1920, we forgot how to make nice buildings. European cities tend to have more construction from before the Great Forgetting, and it makes the built environment much more pleasant.
• Pedestrianized streets. Often with cobblestones.
• In general, European cities are just more pleasant. Given how hard it is to build a good city (or indeed to retrofit one), this feels like a big deal.
• Cured and pickled food.
• Bread. Obviously varies by country, but it’s generally true.
• Voltage. What are Americans doing waiting so long to boil kettles?
• Beauty in the mundane. I find that you’re more likely to find tasteful touches in prosaic places in Europe.
• Motorway design and signage. Standardized, clear, and easy-to-use. The US is a mess by comparison.
• Bathroom doors. That is, in Europe, they’re proper doors. Why does America make us see others’ feet?
• The clangor of church bells on Sunday.
• Trains. Enough said.
• Pharmacies. I'd love to understand why they're so much nicer in Europe.
• Cheese. Again, lots of cross-country variation, but true in general.
• I'm not sure why, but European regulation on many everyday items seems better. Sunscreens in Europe are better, as are bike helmets.
• Wine.
• Languor, joie de vivre, hygge, gemütlichkeit, craic. I think Europeans are better at unwinding. Drawing contrast with what he found in the US, De Tocqueville observed that in Europe "idleness is still held in honor". This difference remains apparent.
• Road density. Europe generally has many more roads per square mile, which makes it easier to find nice places to run, walk, and cycle.
Better in the US
• Air conditioning. Consistently bad in Europe. (Partly for silly degrowth-related reasons?)
• Coffee. Opinions will differ, naturally, but third wave coffee has seen much more enthusiastic adoption in the US.
• Cookie banners. That is, the lack of them. (Well, there are some, but it’s not as bad as the fusillade one is subjected to in Europe.)
• Internet speeds. European wifi often reminds me of my dialup youth.
• Capital markets. If you need money (as a consumer, a small business, or a startup), it’s much easier to get it in the US.
• Being able to buy groceries on Sunday. Inexplicably challenging on the continent.
• Showers. Like the tepid air conditioning, daily ablutions in Europe are conducted beneath parsimonious trickles.
• Urban air quality. Maybe surprisingly, it is, on average, better in the US. The unpleasant whiffs of diesel exhaust is part of the reminder that one is back in Europe.
• Government efficiency. In general, things happen faster in the US.
• Labor laws. As covered in Stripe's annual letter this year, people are more likely to work in high productivity sectors in the US (and thus to earn more). Rigid rules impede this reallocation in Europe.
• Culture of general aviation with many thousands of small airports. There are around 700,000 pilots in the US—far more than there are in Europe.
• Hospitals. A controversial claim, perhaps, but I find that those who have received care in Europe and the US prefer the US.
• Beer. The microbrewery revolution of the US means that it’s clearly the better place for it.
I’ve been surprised by this on a number of occasions. Basic questions about their APIs results in answers containing models and parameters that are ~years outdated, or they claim that features released a few months ago don’t exist. In Cursor, while fixing something unrelated, it will causally change a working integration to be broken.
@ericzakariasson It would be useful to store the composer history per GitHub branch, to resume with the correct context when toggling between branches.
@ericzakariasson it would be great if @cursor_ai could close the loop on frontend development by loading pages under development in a chrome browser to consume error logs and validate layouts/CSS against different screen sizes.
PROMPTS ARE TINY PROGRAMS
We’re now about 18 months into the AI revolution. One thing that was uncertain in late 2022 was whether prompt engineering would be around to stay, or whether better AI would quickly obviate it.
I now think it’s around to stay and I have an explanation that makes sense to me, at least: prompt engineering is just a subset of software engineering.
That is, prompts are tiny programs written in natural language. But the API isn’t specified and varies between models. So guessing the right “function calls” with clever use of vocabulary is a huge part of the game. On the other hand, even if you don’t guess *exactly* the right words to use, the model will often do what you mean.
This is different from how we normally think of an API, which is both more legible and more fragile. The exact words to make an API do what you want are written down, but if you don’t say those exact words it won’t do what you want.
Even given this difference, the concept of prompts as tiny programs using hidden APIs helps explain the bizarre magic associated with specific phrases. I’m reminded of Quake3’s fast inverse square root[1,2], which has a famously obscure incantation in C that just so happened to deliver a 4X speedup.
More code now looks like that, and it makes sense. C is how you talk to machines and English is how you talk to humans. So, just like you write part of a large application in C for performance, you’ll also write part of it in English for dealing with unstructured data.
You can go further with this analogy. Once you think of prompts as code, you can probably generate model-aware syntax highlighters for favored keywords. You can maybe automatically generate API-like docs from a model for the most common use cases.
And you can think of every new model you add to your codebase as roughly analogous to adding a new programming language — because just as it takes time for someone to ramp up on the idioms of Rust, they’ll need to play around with the latest Mistral to get the hang of how to talk to it.
Anyway — this is all probably obvious to folks spending 100% of their time in the field, and is similar to some of the things @karpathy has posted about, but at least for me it was a useful articulation of why prompts are around to stay: prompts are tiny programs.
[1]: https://t.co/3Ryz3N69hY
[2]: https://t.co/qK2C7IU2g1
Overheard in Silicon Valley: "My biggest lesson from the last year is just, it’s all going to happen. All my childhood dreams of space colonies and brain interfaces and AI and robots. Everything is just… going to happen."
I left Brex a couple of months ago to join @RattrayAlex and the team at @StainlessAPI to build the future of REST. Our SDK announcement today tackles the most salient problem: type safety. I'm incredibly proud of what the team has shipped and we're just getting started!
@mercury I dread every time I have to set up a wire to fund an investment with Chase. If the UI allows me to copy & paste an account number, I’m sold.
@karrisaarinen@linear It would be great to get summaries of what every team shipped during the past cycle, especially if it would include information from the linked PRs.