RSSAmplifier

Blog

Daniel Hussey's Blog

About I write code, study medicine, and think a lot about the future. I live in Sydney, where I split my time between hospital wards and terminal windo...

danhussey.bearblog.devRSS feed ↗5 posts

Latest posts

Craftsmanship, per Million Tokens

Artists and engineers seem to be at different ends of the debate on the use of AI in their fields. I see a lot more open contempt for AI from the artistic community than I do the engineering one, surprisingly so. The art world craps on good AI art because it’s AI, undermining their own credibility in the process. EDIT 14/05/2026: A week after I published this, X user @SHL0MS got hundreds of…

Lumine: How a 7B Model Plays Genshin Impact for 5 Hours

My 'oh shit' moment in AI, the moment that its capabilities had me annoying my non-technical friends, was when the OpenAI Five (agents) beat the (human) world champions in Dota 2, back in 2019. Before that, AlphaGo was impressive but kind of academic and distant, like a chess solver. But to this day, generalist real-time gaming agents are still pretty lacklustre. So, when this paper 'Lumine' came…

Durable Advantages in the AI Era

I've always been a tech guy, and I thought one day I'd build and run a tech company – ideally in something cool and futuristic. We're finally getting the futuristic stuff, but I'm not sure the idea of a tech startup works anymore, all for kind of the same reason: AI. This probably only applies to software. Hard-tech startups face different dynamics. I once thought I'd end up in space tech – I did…

I Made Claude Use Design Patterns and It Wrote 11x More Code

What happens when you tell an LLM to use design patterns "liberally"? I ran a quick experiment to find out. The Setup Same model (Claude Sonnet), same problems, two different system prompts: Simple mode: "Write the most direct, straightforward solution possible. Avoid abstractions unless absolutely necessary. If it can be done in 10 lines, don't write 100." Pattern-brained mode: "Use established…

Debugging the ChatGPT Mac App's Keyboard Input Lag

A deep dive into SwiftUI performance profiling and AttributeGraph pathologies I'd finish typing a sentence and then watch the characters continue appearing for another few seconds, my fingers idle while the app caught up. The ChatGPT Mac app had developed a noticeable input lag problem, and I wanted to understand why. What followed was a few hours of profiling, hypothesis testing, and ultimately…