Erik

Software Engineer | Frontend, Full-Stack, Automation

[Erik @ej_dev1]

@NetworkChuck I love this type of posts where you share "life advice" rather than just tech (I love tech too). Keep doing this man!

[Erik @ej_dev1]

@ChShersh Who's in a better position? That's the question.

[Erik @ej_dev1]

@ThePrimeagen Even if you took money, what would be wrong with that?

[Erik @ej_dev1]

@dhh @ThePrimeagen @digitalocean @AIatMeta So happy to hear this!

[Erik @ej_dev1]

@thekitze I honestly thought I would be sad when I stop writing code by hand, but I'm actually happier than ever.

[Erik @ej_dev1]

2026 may be the year of linux desktop, but I heard that 2027 will be the year of htmx desktop.

@htmx_org can you confirm?

[Erik @ej_dev1]

@anthdm It's a classifier, but much better classifier. I had a need for classifiers in my apps dozens of times, but I often didn't use them because I didn't have time/resources to train them.

This is a great tool imo.

[Erik @ej_dev1]

This looks promising 👀

[Diogo Almeida @CompleteSkeptic]

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?

I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

[Erik @ej_dev1]

@terminaldotshop My goal now is now to train hard enough to be able to do this.

[Erik @ej_dev1]

@ThePrimeagen Yeah, but it has less servers. Get it? ServerLESS.

The opposite would be servermore

[Erik @ej_dev1]

I want to write, I want to get good at writing and speaking.

However, whenever I try it feels so hard, I suck at it. I know that I will get better if I do it consistently, but it's so hard to make myself do it which is super weird since I actually want it.

[Erik @ej_dev1]

@dhh Are you interested in https://t.co/ytcWGMiRtd?

[Erik @ej_dev1]

@thdxr Thank you for not making me read

[Erik @ej_dev1]

@ChShersh Is solving solved?

[Erik @ej_dev1]

@thekitze We will call it dumb the minute a new model drops, and people will still build todo apps with it.

[Erik @ej_dev1]

Even before being completely disabled, Fable showed us that a weak uncensored model is much more useful than strong censored model.

[Erik @ej_dev1]

You all may claim that Fable 5 is (or now was) a game changer, something amazing, 100x better than any other model yada yada, but the truth is, the story repeats for every new model.

When Fable 5.1 (or whatever) drops, people will be claiming that Fable 5 is unusable garbage.

[Erik @ej_dev1]

"It's like you're always on your first day, but you're given the task of refactoring the universe."

Yes, I encounter this a lot.

[ThePrimeagen @ThePrimeagen]

I think one thing that is understated is that the hard parts can feel harder with more delegation.

Not only are you attempting to solve hard architectural ideas, you also are constantly solving it in a project in which you don't have the same deep familiarity that we did say a couple years ago.

It's like you're always on your first day, but you're given the task of refactoring the universe.

At least that's something I have noticed that makes me feel a bit angsty

[Erik @ej_dev1]

You can squeeze much more out of LLMs than you think, seriously.

Push them hard and you will be surprised.

[Erik @ej_dev1]

Imagine not learning new things because of AI. Dumb.