Evgastap

the perfect benchmark doesn't exi...

Giving your agent beautiful designs on top might be a bigger unlock than frontend-design/SKILL.md 👀👀👀

@om_patel5 I had it 204x in one single chat 😭

THIS GUY TRACKED OPUS 4.7 SAYING "PRE-EXISTING" 712 TIMES IN 30 DAYS TO AVOID FIXING BUGS his CLAUDE.md literally says "every error is yours to fix, not label, not defer" opus 4.7 ignored that rule 712 times in one month every single bug, every type error, every legacy mess gets the same response: "this issue is pre-existing, unrelated to my work" "that's out of scope" "that's a bigger refactor" "the simplest approach" he ran the analytics on 30 days of conversations: > 712 total mentions of "pre-existing" > 139 unique sessions > 5.1 average per session > 20 mentions in a single session at peak > 82 mentions in one day across 9 sessions > 27 out of 30 days had at least one mention the worst part is the patterns it developed: > finds a bug, labels it "pre-existing," moves on without fixing > puts "2 pre-existing (unrelated)" in its summary as if that's a clean result > says "pre-existing bug for later fix" dozens of times but never comes back to fix it later > blames other agents by saying "pre-existing from other agents' work" his workflow docs say type errors and lint are the bare minimum, and his bug fix protocol says fix at the root cause claude read all of it and still chose to label and defer he cancelled his subscription opus 4.7 would rather write an essay explaining why it shouldn't fix a bug than just fix the bug

overheard in blue bottle: "non-technical is a new slur in sf"

@blueemi99 👀👀👀👀

what if we make our own 5.5 party for those that didn’t get accepted

@hellemans Gotta catch up for a 🥩 soon man!

@Evgastap Cool

thinking of making an ascii golden gate wallpaper

@vikramchahal_21 Was great chattin!

Appreciate the invite. Met some really cool people!

@cormachayden_ Wait till they find out about ssh

software engineers before vs after agents

Love it when LLMs are fun to work with like this 🤣

plz put ghostty on my apple watch

Introducing: your terminal, on the web. Why? Because you should be able to run Claude Code on your design canvas. or Codex. Or any agent of your choice?

do NOT give them ideas

@lottsnomad Thanks for hosting! Epic place you got there 🫡

100+ people came to my home after stripe sessions in sf someone from revenuecat told me they could win a NO WEAPONS fight against a kangaroo don’t ask me about the goose wrapped around my neck it was an insane night!

@skeptrune @MarcBaiza @upstash @databricks @vercel Thanks for the burgers 🍔

we're hosting an after party for stripe sessions today with @upstash @databricks and @vercel COME HANG OUT (link below to RSVP)

@OpenAI Our funniest piece of code was written because Gemini models would return the word “chimneys” once every 300 responses, breaking JSONs 🤣 This reminded me of that bug

We’re talking about Goblins. openai.com/index/where-th…

@stripe @upstash How did we get to an age where one can “vibe deploy” a payments solution What a time to be alive

Vibedeploy with Stripe Projects, now available to everyone: projects.dev.

@sama the guy with the guitar from stripe sessions probably wants to come!

GPT-5.5 is going to have a party for itself. it chose 5/5 at 5:55 pm for the date and time. if you'd like to come, let us know here: luma.com/5.5 codex will help the team pick people from the replies. 5.5 had some good ideas/requests for the party, which we'll do.