Vince Picone

This shit gives me goose bumps. Embarrassing tbh.

@benhylak lol they edited it.

"That kind of A isn't a minor annoyance. It's a B."

@benhylak My body can’t take another quiet disconnect.

@bachiitter @initjean Love experimental using features in production 😩

@initjean this helps? tanstack.com/start/latest/d…

tanstack start would probably be great if it didn't try to bundle server code from API routes on the client-side

This is peak

@cnakazawa Not another quiet disconnect 😭

I think I'm at a breaking point with LLM text. ChatGPT's language has become the worst. I have full on AI fatigue. The honest truth Why this fixes it (short answer) Clean fix The safest bet Final honest take Best-case scenario (totally possible) My straight recommendation Bottom line (no sugarcoating) If you want, tell me […] and I’ll tell you what I’d personally do in your exact situation (not generic advice) Instead of asking “[…]”, think: 👉 “How do I maximize […]?” My honest recommendation (based on what you said) Let me be real with you upfront Here’s the pro move That’s actually a really good question—let’s sharpen it so it actually makes sense. Still real. Not peak performance That’s not just […]. That's […] I wrote the first 3 myself, but then I went to a chat and just kept copying more examples. People don't write like this. Are we doomed to have to read the same poor sentence structure and wording for the rest of our lives? It's even worse when I have to read other people's llm slop. Thank you, I can prompt an llm myself. Do I have to pay a person to operate the llm for me and write back slowly in human language?

@zachkrall is this a linear subtweet.

remember when you would log into an app and you were presented with a dashboard that was maybe kind of useful to you and now you just get an empty page with a chat input

@KalessLW @SapphoSys mrshu.github.io/github-statuse…

@SapphoSys not what I see on githubstatus.com

world's first enterprise solution to reach zero nines uptime

@tmuxvim Not everything needs to be practical. You can just build interesting things.

why are people so obsessed with this lol like it's cool tech but almost every demo I see of it is totally impractical. it should just be used for laying out articles in more intersting ways

"It makes the text unreadable!" Yeah we know.

The pretext discourse in a nutshell.

@matthewcp It’s one tap for _you_ to send the email. It takes more taps for you to type out the request to send it.

@vppicone How can it send an email?

Can the anti-MCP folks explain how to do anything non-devtools with their method? For example, I want to send an ebook to my reader from my phone. What am I supposed to do, fire up a remote container to run my vibe coded CLI?

@matthewcp Why does the thing that’s generating audio books also need to be the thing that’s sending them. This is the MCP bloat incarnate.

@matthewcp If you have a Claude generate these, It’s a 2 button share from their app to your device. Spinning up an MCP service for that seems wild.

@matthewcp Depends on the device. There’s a dozen ways that don’t involve spinning up a server to instruct the AI on what to do.

@matthewcp Kindle lets you email mobi/epub files to a kindle email. I’d simply have it do that.

@vppicone Great, what's 1 way other than MCP? Also why does it depend on the device? Let's say you're using Claude or Gemini or ChatGPT via their mobile apps. How do you do it?

@GeoffreyHuntley Come check out NYC Resistor, it’s a hackerspace in Gowanus. We have open nights Monday’s and Thursday’s, lots of cool attendees.

so; what things should i put on my bucket list for nyc? last time i was here was circa 15 years ago.

Today we're releasing our first open source TTS model, TADA! TADA (Text Audio Dual Alignment) is a speech-language model that generates text and audio in one synchronized stream to reduce token-level hallucinations and improve latency. This means: → Zero content hallucinations across 1,000+ test samples → 5x faster than similar-grade LLM-based TTS → Fits much longer audio: 2,048 tokens cover ~700 seconds with TADA vs. ~70 seconds in conventional systems → Free transcript alongside audio with no added latency

@gerardsans It’s hilarious that this coincides with defaulting the model to moderate effort as well. Wonder how many of the issues would be avoided with high effort.

Make it make sense, Anthropic: • Claude writes a PR → bills you for the “solution” • Sixty seconds later, Claude sees its own code on GitHub → bills you again to review its rookie mistakes Next: pizza arrives cold → extra charge to “inspect”? Classic double-dip. Chef kiss

@trq212 @Miles_Brundage Web client, not cc

@Miles_Brundage like the model in /model is just sonnet 4.6 now and you don't remember changing it? definitely a really weird bug, we would see it at scale if it were affecting a lot of people but possible it's a more specific bug that happens rarely... will look into it

Lately, Claude has been defaulting to Sonnet in a way that I don't think it ever did before. PLEASE STOP THIS, IT'S REALLY ANNOYING