Tessera
Folgen

Andrew Duffy

@a10y.dev

SYSTEMS HACKERS SOLVE THE BEAR MENACE

161 Posts214 Follower275 Folgt
Andrew Duffy@a10y.dev · last mo.

i'm in the trenches. i'm at war. i'm in the arena. i'm an operator. i'm a killer. i move mountains. we're winning. we're revitalizing. we're reindustrializing. we're the shield of western civilization - guy doing LLMs for procurement

Andrew Duffy@a10y.dev · 3mo ago

Just burned a couple hours debugging a handheld Pi project, seeing HAT not getting power. Then got out my probe and ran continuity test. "hmm that's weird, why are all of these holes sharing a node..." I ordered stripboard instead of perfboard 🤦

Andrew Lawrence@ndrew.bsky.social · 5mo ago

billionaire being interviewed: i havent listened to a single note of music in 27 years. think about it, music was basically invented, what, maybe 50 years ago? interviewer: this is really interesting

Andrew Duffy@a10y.dev · 5mo ago

It's still hilarious that when I ask Claude to draft up a plan for a complex task, it gives me human-time estimates. "Part 1: Months 1-3. Part 2: Months 4-6" But then you just tell it to go and it finishes everything in a few minutes lol

Andrew Duffy@a10y.dev · 6mo ago

Look, I get it, Lobsters is sort of the un-HN. Which is cool, I've been a lurker/occasional submitter for over a decade now. The vibecoding tag drama is annoying though. Coding agents are obviously technically/culturally interesting, the repeated attempts to censor it seem misguided

Andrew Duffy@a10y.dev · 6mo ago

Perfect snow day activity: making a toothbrush stand for our small bathroom. Part 1: designing the support bracket and doing a test print

Simon Willison@simonwillison.net · 7mo ago

This is great - it's about time someone updated the discourse on LLM energy usage to reflect that coding agents use massively more prompts than occasional questions to ChatGPT Simon estimates that a day of coding agent usage comes out close to the energy needed to run a dishwasher

Simon P. Couch@simonpcouch.com · 7mo ago

Whenever I read discourse on AI energy/water use that focuses on the "median query," I can't help but feel misled. Coding agents like Claude Code send hundreds of longer-than-median queries every session, and I run dozens of sessions a day. On my blog: www.simonpcouch.com/blog/2026-01...

A bar chart showing the electricity use of several daily activities with the subtitle "The 'typical query' is not a useful way to think about coding agents' energy use." The bar for a 'typical ChatGPT query' is not even visible. My median Claude Code session is somewhere between the average US household per minute and toasting bread for three minutes. My median day with Claude Code is something like running a dishwasher.