Kieran McHugh

@kmc.bsky.social

Founder of @daybridge.com, @monzo.com before that Software engineer, designer, and piano player 🇬🇧 🏳️‍🌈 London, UK @kieranmch formerly on X

I love Claude Code and Opus is incredible, but man they really should stop bragging about it being vibe coded - it's a damning indictment of vibe coding. By far the most glitchy AI software I have used.

Bild

I'm trying to understand why the profiler says a component is re-rendering because 'the parent component rendered' when the parent definitely didn't render? It's wrapped in React.memo and none of the props are changing (or I presume it would say so?) Paging @ricky.fm @danabra.mov 🙏

BildBild

I love living in London, I’ve been here for nearly 10 years. It might be the best city in the world. The far right hate it because it’s living proof that multiculturalism works and that different cultures can coexist and thrive. Trump and DHH hate it because it disproves their narratives.

Jake Lazaroff@jakelazaroff.com · 11mo ago

show this to anyone else you know who also assumed DHH was just a normal conservative dude rather than a raging white supremacist lunatic

I mean, our brains do the same thing only with a biological substrate instead of a silicon substrate. LLMs are “probability machines” but their output is based on a real, acquired understanding of the world from a training data set. Different to humans but still arguably thinking.

Gergely Orosz@gergely.pragmaticengineer.com · 12mo ago

Amusing how 99% of people using LLMs forget how these things work: They are advanced probability machines. They generate the next most likely token (word) based in the input and their training. Under the hood, it’s a giant matrix multiplication that has eerily good output.

In the UK we should do anti-honours. And the first person to get an anti-Knighthood should be whoever removed all of the fucking padding from the seats on the trains and replaced them with ironing boards.

Biggest red flag for AI generated text is excessive use of present participles 🚩 "highlighting the importance of", "underscoring the relevance of", "avoiding the risks of", "emphasising the need for" etc

Lmao StackOverflow mods are going through and closing questions I asked 10 years ago. They shall not rest until they have eliminated every question on the platform.

I’ve seen an uptick in people on LinkedIn posting pictures of sick relatives and using things like relatives’ cancer diagnoses for content. It’s kinda nauseating tbh.