[KaiMagnus]

I got access to Kiro from Amazon this week and they’re doing something similar. First a requirements document is written based on your prompt, then a design document and finally a task list.

At first I thought that was pretty compelling, since it includes more edge cases and examples that you otherwise miss.

In the end all that planning still results in a lot of pretty mediocre code that I ended up throwing away most of the time.

Maybe there is a learning curve and I need to tweak the requirements more tho.

For me personally, the most successful approach has been a fast iteration loop with small and focused problems. Being able to generate prototypes based on your actual code and exploring different solutions has been very productive. Interestingly, I kind of have a similar workflow where I use Copilot in ask mode for exploration, before switching to agent mode for implementation, sounds similar to Kiro, but somehow it’s more successful.

Anyways, trying to generate lots of code at once has almost always been a disaster and even the most detailed prompt doesn’t really help much. I’d love to see how the code and projects of people claiming to run more than 5 LLMs concurrently look like, because with the tools I’m using, that would be a mess pretty fast.

[peanutz454]

> At first I thought that was pretty compelling, since it includes more edge cases and examples that you otherwise miss.

So, is it good for writing requirements, and creating design, if not for coding?

[scuff3d]

I doubt there's much you could do to make the output better. And I think that's what really bothers me. We are layering all this bullshit on to try and make these things more useful then they are, but it's like building a house on sand. The underlying tech is impressive for what it is, and has plenty of interesting use cases in specific areas, but it flat out isn't what these corporations want people to believe it is. And none of it justifies the massive expenditure of resources we've seen.