floriferous

This is so amazing. Thank you for building this. I hope you keep refining it and making it better for different use cases!

People need to actually put in some effort with LLMs here. Ask any SOTA agent to identify a specific type of "janky crap", and it'll quickly find every single occurence. Fixing it with the right pattern should be trivial for an experienced engineer, and therefore its agents.

Just do this for every type of "janky crap" and your codebase converges to perfection.

It's not that hard, but it takes a lot of tokens, and setup.

Let's be honest, their model is not good enough for them to get any ROI on it if it was closed source.

They're making it look like a benevolent decision, but they really don't have another option.

Is this comparable to Supabase's just announced multigres?

I think openAI and Anthropic are getting ready to launch their next generation of models (Claude 5 and GPT 6), which will quickly make us hit our rate-limits and we'll start entering a world where most people will start to have/want to pay for additional tokens.

We're all just getting too used to having great models for a fraction of the the value they give us.

While this is really concerning, it feels like a small new category of errors to check for. The article mentions an increase of 220% of the amount of submissions. That's incredible news for science, probably lots of honest scientists able to produce more work and eventually lead to more science being done.

Agreed, Sentry, Posthog, and many more are all doing the exact same thing now, I'd be surprised if this was a good deal for Langfuse. I personally migrated away from it to use Sentry, their software was honestly not that great.

Exactly the same feeling here, it's such a strange thing to promise a kid they are accepted into a program, but it starts in like 4 years.

As if all of your education (and probably side-projects, life decisions) is meant to lead to YC.

Great metaphor, exactly how it feels to me too!

The blinking thing after the author's name in the sidebar is terribly distracting, why make it impossible to scroll past it, or just not have this infinitely running animation? Great article otherwise.

This is such an insanely good and important idea, that it feels inevitable that we'll eventually do it, probably in the next decades/centuries.

Probably one of the hardest new laws we'll ever have to implement.

Modern Gaming feels like a chore compared to older games. It isn't just nostalgia. Most games are designed to waste your time for maximum player engagement. Modern Gaming is boring by design.

I just can’t function optimally without 3 monitors:

1 for my full screen IDE: 3 files side by side, multiple terminals

1 with 2 browser windows: my product (a web app), and another with specs, stack overflow, google

And my macbook’s screen with slack or email.

Whenever I have 2 monitors or less, I find myself switching windows around all the time! With 3 I just move my eyes and mouse.