Sabitlenmiş Tweet
Having trouble getting Claude to one-shot the thing you're trying to do? Have it one-shot a tool that lets you do the parts it is struggling with instead.
English
Tyler van Hensbergen
1.1K posts
Tyler van Hensbergen
@tvanhens
Engineering @fluidstack
Vancouver, WA Katılım Eylül 2013
413 Takip Edilen518 Takipçiler
Bend2's standard library is now racing from having ~10 functions to being larger than Haskell's Prelude, except better organized and formally verified. All these functions are trivial to modern AI's.
But then, this raises the question: why do we even need libs anymore?
English
Tyler van Hensbergen retweetledi
English
@kenwheeler The art of the iteration is where all the fun is, plus you get better results that way too. Can't have happy little accidents without being in the loop.
x.com/tvanhens/statu…
Tyler van Hensbergen@tvanhens
Having trouble getting Claude to one-shot the thing you're trying to do? Have it one-shot a tool that lets you do the parts it is struggling with instead.
English
i feel like a lot of the art here is in the iteration. you try shit, decide it’s cool or pivot. repeat. i just don’t understand what kind of shit can be so tightly defined in its entirety that it’ll just run for 3 days and you get what you want
English
i’m curious what you guys actually have agents build 24/7. i don’t have that many plans. is it just the dumbest shit ever? do you stop and think for a while about something to build?
English
Tyler van Hensbergen retweetledi
SemiAnalysis x Fluidstack is kicking off GTC with a Power to Prefill, Dirt to Decode, Transformers to Transformers: A Full-Stack AI Infrastructure Hackathon.
Build with the best. Bring your project, win prizes, and listen to insights from top industry leaders like @marksaroufim GPU MODE, @cHHillee Thinking Machines, Thomas Raoux OpenAI, @dylan522p, and more.
Limited spots, apply below. You won't want to miss it.
luma.com/SAxFSHack
English
@Ronalfa The real AI is the friends you made along the way.
English
@Ronalfa @iScienceLuvr Gottah get your clawbots going then you can prompt while you romp.
English
"oh, it's valentine's day? i totally forgot i was too busy grinding on my B2B SaaS startup" - half the founders on Twitter
English
@thdxr Would love to see how this one is going in 6 months.
English
every time we talk about how much companies are spending on LLMs per engineer we get a bunch of replies "wdym claude max is just $200"
when rolling out coding tools to a real team companies use a proper enterprise control plane attached to a yearly contract
these all have a pay per use component (with discounts) so yes it's not completely extraordinary to see a $2000 bill for a single dev per month
and no just because $2000 is less than the salary doesn't mean it's automatically worth it
eventually companies will scrutinize these numbers and understand if they're actually seeing proportional ROI
English
Developers that have an attachment to how the code looks are in for hard times.
English
Doing this to bootstrap your own product is dumb. But doing this on an individual basis to shape the tool to how you need to work feels like the inevitable end state.
dax@thdxr
there's a company that forked opencode and are posting as though they built it from scratch idc about all that but they couldn't even find and replace things right and i had to submit a PR to them to correct their user agent saying "opencode"
English
@tlakomy Last picture is me using cursor just as a shell for Claude code.
English
@thdxr Said another way: LLMs are great at doing what has been done before.
English
@thdxr The ideal for me would be only loading the closest and then having an easy way to load additional ones from your context file. Make load all opt-in but easy to opt-in. Context is precious, overloading it degrades results.
English
We’ve only been able to get our agent to generate slop, so you’re going to have to eat it out of a trough.
ry@rywalker
software development in 2026 is going to require some to loosen up a little code doesn't have to be as perfectly crafted the way we did it pre-ai call it slop if you want, but if you're still demanding perfection on every pr while your competitors are shipping "slop" that works... you're fighting from a disadvantaged position shipping velocity matters more than perfection
English
Tyler van Hensbergen retweetledi
Today we’re making New York City Fluidstack’s global HQ.
We’re already delivering the AI infrastructure the U.S. needs to operate at speed, and we’re excited to relocate our headquarters to the place that has quickly become our center of gravity.
English