Shame but also opportunity... These apps will not be good, they will look good and appear to work, but then... they will need to be rewritten.

I am often posting that if you pick Gleam for UI, Rails or Phoenix, which will make your codebase easier to follow and more stable, I will give you discounts.

Gleam as a language is genuinely underrated for porting.

I had made a Golang htmx + templ codebase and just out of curiosity I ported it onto gleam using an open weights model (glm 5.3)

Although it struggled a little bit first because of learning some parts of gleam but overall after it searched for documentation and learnt some things on its own, it was able to completely rewrite it in just 1/2 very small prompts which is crazy.

I think phoenix can be great as well and I am doing an experiment to try to port it to many functional languages and just testing which language is the best for it but so far I am impressed by gleam!

Yeah my agents do spend a little bit more time with it, and I use GLM Flash mostly. Gleam has good error messages which helps a lot. But once they write something it is much more solid.

Yea, I think that for AI, good error messages might just be worth their weight in gold.

I was just trying to port the same website to elixir and I am finding that there are issues within it and also for some reason the agent got stuck at something and then it started writing elixir within the repl and doing something related to ecto.

In my current experience, contrary to what I would've expected (given that elixir is a more mature language than gleam), gleam is surprisingly better, though I will try to do more tests

Some other thoughts that I am having at the moment is that starting a project in gleam or within sveltekit/astro/solidjs and then porting it to gleam can be great if what you are doing is frontend heavy. AI will take time once to learn gleam but then the rewrite becomes much more solid and personally interesting to me

Looks like I have to learn gleam to see how fun & managable the codebase can be :-D

Legitimate question. In my projects, where I do have a ton of files, but kind of suspect not nearly as you have, I do occasionally instruct agent to review old files and decisions and archive them or delete.

Sometimes it is useful to know how we came about some decision or feature, but in general you often don't need them and they would contain prose that would confuse agent. So you remove it.

I don't use Claude Code mostly because it always loved to generate reports and there is no way I will read it's lengthy reports. I try to have documents I can actually read if I want to.

Hope this helps, but some lifecycle of your documents and requirements is useful to have.

I will just say, as of recently, especially thanks to AI, I've been doing projects in Gleam, I am on third Gleam project and it has been very pleasant experience. It takes longer to write but it is much shorter, easier to follow and much more easier to update and change.

So all we have to do is send telescope right up and let it record...

I will just say that this short movie is something most beautiful I've seen in last several years. To be able to see this, for real, not as a side-effect to the star is absolutely mind blowing.

I am confused over this... I get efficiency but price seems too high to me. I didn't try the model, so it migth be beyond fast or some other quality that is not obvious.

Anyone knows more or used this model?

AI will do all our work now, but man at least I lived and wrote Pascal growing up.

Spring 2027 is eternity in the future from today. I love the concept they advertise but by then, there will be so many amazing products and releases.

If they could give me this today and worked somewhat like they promise, even though for $500 or so I can get glasses that do what I need them to do today, for more fidelity I would gladly pay 1k+ they are asking.

I am not sure what will happen till Spring or Summer next year.

Now when I saw their video... ufff, not for me.