Claude Sonnet 4.5

> The difference is "negative" people can clearly see that it's on a trajectory in the NEAR, not even distant, future to completely eat your earnings, so they're not thrilled.

We birthed a level of cognition out of silicon that nobody would imagine even just four years ago. Sorry, but some brogrammers being worried about making ends meet is making me laugh - it's all the same people who have been automating everyone else's jobs for the past two decades (and getting paid extremely fat salaries for it), and you're telling me now we're all supposed to be worried because it's going to affect our salaries?

Come on. You think everyone who's "vibe coding" doesn't understand the pointlessness of 90% of codemonkey work? Hell, most smart engineers understood that pointlessness years ago. Most coders work on boring CRUD apps and REST APIs to make revenue go up 0.02%. And those that aren't, are probably working on ads.

It's a fraction of a fraction that is at all working on interesting things.

Personally, yeah, I saw it coming and instead of "accepting fate", I created an AI research lab. And I diversified the hell out of my skillset as well - started working way out of my comfort zone. If you want to keep up with changing times, start challenging.

Well, here's an even more insane claim: I'm infinity times more productive, as I just wouldn't even start projects without the LLM to sidestep my ADHD. Then, when the LLM invariably fucks up, I step in and finish things myself!

Here are a few projects that I made these past few months that wouldn't have been possible without LLMs:

* https://github.com/skorokithakis/dracula - A simple blood test viewer.

* https://www.askhuxley.com - A general helper/secretary/agent.

* https://www.writelucid.cc - A business document/spec writing tool I'm working on, it asks you questions one at a time, writes a document, then critiques the idea to help you strengthen it.

* A rotary phone that's a USB headset and closes your meeting when you hang up the phone, complete with the rotary dial actually typing in numbers.

* Made some long-overdue updates on my pastebin, https://www.pastery.net, to improve general functionality.

* https://github.com/skorokithakis/support-email-bot - A customer support bot to answer general questions about my projects to save me time on the easy stuff, works great.

* https://github.com/skorokithakis/justone - A static HTML page for the board game Just One, so you can play with your friends when you're physically together, without needing to bring the game along.

* https://github.com/skorokithakis/dox - A thing to run Dockerized CLI programs as if they weren't Dockerized.

I'm probably forgetting a lot more, but I honestly wouldn't have been bothered to start any of the above if not for LLMs, as I'm too old to code but not too old to make stuff.

EDIT: dang can we please get a bit better Markdown support? At least being able to make lists would be good!

That's what I was thinking too; the models have the same data sources (they have all scraped the internet, github, book repositories, etc), they all optimize for the same standardized tests. Other than marginally better scores in those tests (and they will cherry-pick them to make them look better), how do the various competitors differentiate from each other still? What's the USP?

I'm on a max ($200) plan and I only use opus and I've _never_ hit a rate limit. Definitely using for 5+ hours at a time multiple days per week.

> include some specifics of the task and prompt. Like actual prompt, actual bugs, actual feature, etc.

> I am still surprised at things it cannot do, for example Claude code could not seem to stitch together three screens in an iOS app using the latest SwiftUI (I am not an iOS dev).

You made a critical comment yet didn't follow your own rules lol.

> it's so helpful for meaningful conversation!

How so?

FWIW - I too have used LLMs for both coding and personal prompting. I think the general conclusion is that it when it works, it works well but when it fails it can fail miserably and be disastrous. I've come to conclusion because I read people complaining here and through my own experience.

Here's the problem:

- It's not valuable for me to print out my whole prompt sequence (and context for that matter) in a message board. The effort is boundless and the return is minimal.

- LLMs should just work(TM). The fact that they can fail so spectacularly is a glaring issue. These aren't just bugs, they are foundational because LLMs by their nature are probabilistic and not deterministic. Which means providing specific defect criteria has limited value.