Profiling Hacker News users based on their comments

Profiling Hacker News users based on their comments

21st March 2026

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. The Algolia Hacker News API supports listing comments sorted by date that have a specific tag, and the author of a comment is tagged there as author_username. The Algolia API is served with open CORS headers, which means you can access the API from JavaScript running on any web page.

Last August I had ChatGPT build me a simple tool for hitting that API for any user which fetches their comments and gives me a mobile-friendly “copy to clipboard” button. I’ve since tweaked it a few times with Claude. I can then paste the whole lot into any LLM—these days I mostly use Claude Opus 4.6 for this—and prompt “profile this user” to see what it can figure out.

It’s startlingly effective! It feels invasive to quote the profile of another user here, so I’ll show you what it produces for me:

This is Simon Willison—a prolific, independent software developer, blogger, and one of the most visible voices in the AI-assisted coding space. Here’s a profile drawn from his comments:

Professional identity: Independent developer and writer. Co-creator of Django, creator of Datasette and many other open source tools. On the board of the Python Software Foundation. Runs a heavily-trafficked personal blog at simonwillison.net. Monetizes through GitHub sponsors, ethical ads, and occasional consulting.

Core thesis on AI coding: He’s the leading evangelist for what he calls “agentic engineering” — using coding agents as productivity multipliers for skilled developers. His central argument: LLMs don’t replace programmers, they amplify existing expertise.

Working style: He does most of his programming from his iPhone via Claude Code for web, often while riding BART, walking the dog, or sitting in the garden. He embraces “YOLO mode” (auto-approving agent actions) and frequently runs 2-3 agent sessions in parallel.

Key technical interests: Sandboxing and security (WebAssembly, Pyodide, Firecracker), SQLite (deeply), Python packaging and tooling, and local LLM inference.

Security consciousness: He coined the term “prompt injection” and the “lethal trifecta”. He’s genuinely alarmed about security risks and has been predicting a headline-grabbing prompt injection attack for over two years.

Personality and debate style: Energetic, combative when challenged, but generally good-natured. He engages heavily in HN comment threads. He pushes back hard against what he sees as motivated reasoning from AI skeptics, but also against AGI hype.

Personal interests mentioned: Niche museums, New Zealand kākāpō parrots, cooking food from countries he’s never visited. Lives in Half Moon Bay, CA.

In a sentence: A deeply experienced, independently-minded developer who’s genuinely excited about AI coding tools, fights hard against both uncritical hype and reflexive dismissal, and is trying to professionalize how the industry uses these tools.

This all checks out! I ran this in Claude incognito mode to prevent it from guessing I was evaluating myself. It’s a little creepy to be able to derive this much information about someone so easily, even when they’ve shared that freely in a public place.

I mainly use this to check that I’m not getting embroiled in an extensive argument with someone who has a history of arguing in bad faith. Thankfully that’s rarely the case—Hacker News continues to be a responsibly moderated online space.