Ask HN: Do you have any evidence that agentic coding works?

I had no idea so many people still clung to these lines!

Personally, I see huge value. I’ve built more in the past year than I ever have before. Ideas that used to go on a list, now get implemented in a weekend. The ones that are good, I’ve shipped. Others get played with, tested, and set aside. Not due to lack of quality, but the idea wasn’t right. Some need more thought so I’ll keep testing, and tweaking over time.

Evidence? Those projects would not exist if not for these tools. Not because I couldn’t write them, but because I WOULDN’T have written them.

But I’ve given up trying to convince people. If you don’t think it works, great. Don’t use it. That’s fine with me!

I’d urge you to keep trying though. Hopefully you have the moment as well. And if you think, “it’s ok for some things, but not what I do”, well…just don’t wait a year before trying again. Keep an open mind, and integrate the tools into your workflow everyday. Not as a one off “ok fine, I’ll test this stuff” grumpy kind of way, but in an honest iterative way where you are using the tools on a daily basis, even in a small way. For docs. Whatever. Just keep using them. Eventually you’ll see. Eventually you’ll have the moment.

One thing I think is lost in a lot of these conversations is how much FUN I am having. I’ve been coding for…well if you count copying BASIC from books into a Timex Sinclair, about 40+ years. Maybe professionally for 30+? I haven’t had this much FUN building since back when I copied those games from that little red book of BASIC programs. These tools make even mundane work fun for me, cause you need to have your workflow right. So coding becomes more like Factorio or something. Keep optimizing your setup, then simplify, the optimize, then simplify.

Syntax was always something that felt like a hurdle. I always rolled my eyes at people that get lost in the minutia of a language. Like I remember the Perl neck beards from way back. lol.

I learned what I needed to get the task done, then moved on. Likely forgetting 90% of the details anyway, so I could keep an eye on what I considered important. Architecture. Functionality. Separation of Concerns. Could I dive back in and figure out the details again if needed, sure. No problem.

NOW these tools let me work at this abstraction level even more.

And I’m here for it.

Just amazing this is where we are at as an industry.

Exploratory scripts, glue code—what I think of as digital duct tape between systems—scaffolding, probes, and throwaway POCs have always been messy and lightly governed. That’s kind of normal.

What’s different now is that more people can participate in that phase, and we’re judging that work by the same norms and processes we use for production systems. I know more designers now who are unafraid to code than ever before. That might be problematic or fantastic.

Where agentic coding does work for me is explicitly in those early modes, or where my own knowledge is extremely thin or I don’t have the luxury of writing everything myself (time etc). Things that simply wouldn’t get made otherwise: feasibility checks, bridging gaps between systems, or showing a half-formed idea fast enough to decide whether it’s worth formalising.

In those contexts, technical debt isn’t really debt, because the artefact isn’t meant to live long enough to accrue interest or be used in anger.

So I don’t think the real question is "does agentic coding work?" It’s whether teams are willing to clearly separate exploratory agency from production authority, and enforce a hard line between the two. ( I dont think they'll know the difference sadly) and without that, you’re right—you just get spaghetti that passes today and punishes EVERYONE six months later.

Most people understanding "hacking" differently than us, but we've made that work, we can talk about hacking here without other HN users believing we're cracking passwords, why not the same for other terms?

And that problem is not just between people, this lack of communication skill continues with people's internal self conversations. Many a bully personality is that way because they bully themselves and terrorize themselves.

It's no wonder that people can use AI at all, with how poorly people communicate. So the cluster of nonsense that is all the shallow thinkers directing people down incorrect paths is completely understandable. They are learning by doing, which with any other technology would be fine, but with AI to learn how to use it by using it can serious damage one's cognitive ability, as well as leave a junkyard of failed projects behind.