ukuina

Neat!

Are there any tools that tolerate slight mismatches across lines while combining them (e.g., a timestamp, or only one text word changing)?

I attempted this with a vector DB, but the embeddings calculation for millions of lines is prohibitive, especially on CPU.

Sounds list: {Balanced, Bright, Dark} Noise, Ocean, Rain, Stream, Night, Fire, Babble, Steam, Airplane, Boat, Bus, Train, Rain On Roof, Quiet Night.

They're pretty high quality!

> Son, your system is not durable, your tests are misleading, and you can't reason about what's happening

You just described the majority of pre-LLM enterprise software.

Awesome platform.

I wanted to upgrade!

But your "upgrade to Pro" button on the Account page gets stuck on "Processing..."

This is why other "AI browsers" that parse and simplify the DOM, then invoke a tool-calling LLM over text are at EOL.

Once Chrome integrates Gemini Live amd treats your browser as a video input stream, it's pixels all the way. No lag, no incorrect clicks on hidden elements.

[June 2025]

You will ALWAYS want to use the absolute best model, because your time is more valuable than the machine's. If the machine gets faster or more capable, your value has jumped proportionally.

> While we’ve implemented protections, they aren’t full proof.

Nothing is.

This is big news if it means realtime subtitle generation.

Not sure if you're being serious here, but coining the term doesn't make him responsible for the other things you've mentioned.

OpenAI Codex is available on the mobile app.

The fact that o4-mini coexists with 4o-mini is... a choice.

"I see the problem!"

It generally feels like Opus gives you the 5th or 10th iteration, but Sonnet gives you the first possible solution.

You're going to love MCP.

Nit: The post keeps referring to "standard JSX" as though trying to will such a thing into existence.

At what age?

Is monetization coming soon?

i.e., toggle a "Commercial" switch and users are charged API cost plus 10-30% with the creator getting a third of that?

Also, the switch would disable "See, fork, and customize any artifact"

> I was embarrassed that the release primarily made life better for our developers rather than for our users. Given the tasks I planned for a sprint, I identified the ones that would be exciting and valuable enough to the user to include in the release announcement.

> I thankfully avoided ever writing another uncomfortable explanation of a release that offers nothing to the user.

Seems short-sighted?

This is marketing-driven development where you have "learned the lesson" that thankless maintenance and stabilization tasks should be avoided.

> For complex objects such as computers to exist in our universe, many other objects had to form first, such as stars, heavy elements, life, tools, technology and the abstraction of computing. All this takes time and is path-dependent due to the casual contingency of each innovation that is made.

So... a universal tech tree?

Shh, that'll blow up the hype bubble.

This is why CUAs are going to demolish all prior fences.

ollama run gemma3:1b

https://ollama.com/library/gemma3

> support for over 140 languages

I believe the term is Generative Engine Optimization (GEO).

While this specific app may be paid, I built https://HackYourNews.com (free) specifically to "summarize HN for me" and it grew to include many other websites. Uses Gemini Flash under the hood and summarizes both the story AND the HN comments!

Because it's a distributed for loop?

They handle file locking by themselves. "File1 is currently awaiting edit approval, accept its changes and continue here?"

> so much time is spend waiting for the request to complete (while I work on another part of the app in a different workspace with Cursor).

You can open up to three parallel chat tabs by pressing Cmd+T

https://docs.cursor.com/kbd

Each chat tab is a full Agent by itself!

Aren't most enterprise customers using AzureOpenAI?