Profiling Hacker News users based on their comments

https://simonwillison.net/2026/Mar/21/profiling-hacker-news-users/

Profiling Hacker News users based on their comments

Here’s a mildly dystopian prompt I’ve been experimenting with recently: “Profile this user”, accompanied by a copy of their last 1,000 comments on Hacker News. Obtaining those comments is easy. …

Simon Willison’s Weblog

>
This is arguably their defining HN characteristic: they are one of the most vocal, persistent AI optimists on the platform. They claim ~90-95% of their shipped code is AI-generated, report 5-10x productivity gains, and have built a detailed methodology around it — using Playwright for visual verification, static typechecking as a hallucination filter, and e2e test suites as automated validation harnesses

Wow, I sound really annoying. Sorry about that everyone!

I sound like an annoying old people I guess so I think I'm worse. Either way I forgive you. (GPT called me a wiring closet gremlin)
This just in; posting ridiculous amounts of personal information on the internet can lead to you being profiled correctly. Wild stuff.
Not entirely correctly though, since there are forms of censorship even on HN, which selectively blinds any method of analysis in a systematically biased way.
There was an Irish mafia guy who was caught because his anonymous restaurant review profile got linked to him.
We all know this, but in the past someone probably wasn't going to go through thousands of your comments unless you've really pissed them off. It's worth realising how much lower the activation energy is these days.
Given a profile like this, how good would an LLM be at figuring out whether the profile if from a bot or a real person?

Not doubting the method works in general, but Simon Willison is a public-enough figure so the baseline level of info is higher than just HN comments. If you turn off Claude’s web search:

> Simon Willison is a British software developer, blogger, and open-source advocate, best known for…