daringrain32781

If you want an OTS router then OpenWRT, but make sure it’s compatible first.

If you want a bit more power or firewall capability a minipc running OPNSense also works.

> Write good specs into files, usually Markdown.

When I'm building something new there's a good chance I don't know about a lot of problems (unknown unknowns). How do you overcome this with just spec files?

I keep coming back to this and the only way to develop properly without tons of hacks patching bugs afterwards is by working at a low level. I don't really write code but I work at the code level still and have Claude write each function and whatnot.

> they write compiler perfect code all the time

I wouldn't say this at all. They mess up regularly. What makes them able to write syntax that compiles is their harness which verifies the syntax and provides the feedback loop for them. But LLMs will happily give you code that doesn't compile.

I'm not worried. The rate of change hasn't been that huge. The models still struggle with the same weaknesses from day 1. And if my job was ever threatened, many other fields would be before that.

Exactly. And there is architectural things missed by agents or just wrong. There is equally things I learn all the time from agents doing things, or things I have missed. Together we’re a better team.

For me, price. Bambu's offerings are a fraction of what Prusa charges.

This is happening at my company: all of these software phases are being delegated to AI. And the people without any understanding of software have absolutely no idea what they're doing. Features no one needs are being made in minutes. It hasn't been long enough to really tell how that's turning out. Everything seems to be in endless development.

It’s for developers as far as I understand, it’s not meant to buy as a consumer router. There is far better hardware you can get that runs OpenWRT.

I just use a WireGuard tunnel between my OPNsense router and a VPC instance. Currently using the tiniest t4g AWS instance but it doesn't much matter what the upstream machine is, as long as it has a public IP. WireGuard is easy and I'm not so reliant on a 3rd party service.

Author writes something original, asks the AI to make it sound better, then posts the output of the AI.

Like a lot of things, it’s neither and somewhere in the middle. It’s net useful even if just for code reviews that make you think about something you may have missed. I personally also use it to assist in feature development, but it’s not allowed to write or change anything unless I approve it (and I like to look at the diff for everything)

Seems to me like they’ve decided (or want to believe) that if all code isn’t output by an agent the company is losing out on potential work output.

Agents are still far too unreliable and dumb for this model and need strict discipline by a developer who really understands fundamentals. And sometimes it’s just faster to do the damn thing yourself instead of writing a whole paragraph to an agent that still might do it wrong.

In my view with current LLMs: they still produce far too much bloat and unclean solutions when not targeting them at very specific issues/features, making LLMs essentially a requirement for any debugging or features for the lifecycle of the product/service.