Super HN

New Show
   Claude Code is all you need (dwyer.co.za)
I love this article just for the spirit of fun and experimentation on display. Setting up a VPS where Claude is just asked to go nuts - to the point where you're building a little script to keep Claude humming away - is a really fun idea.

This sort of thing is a great demonstration of why I remain excited about AI in spite of all the hype and anti-hype. It's just fun to mess with these tools, to let them get friction out of your way. It's a revival of the feelings I had when I first started coding: "wow, I really can do anything if I can just figure out how."

Great article, thanks for sharing!

The title is a bit exaggerated. The depth of the projects covered in the article is clearly not representative of "all".

In fact, I now prefer to use a purely chat window to plan the overall direction and let LLM provide a few different architectural ideas, rather than asking LLM to write a lot of code whose detail I have no idea about.

This article feels like it was written as a dialectical exercise between an AI and a human. It would probably benefit from some more heavy human editing to make it more succinct and to give the overall article a structure. As it is, it's very difficult to follow along.
Is Claude Code really that good? I'm currently using Cursor and I let it pick the LLM model to use.
When it comes to diffs (edits), Cursor is batch-oriented, while CC suggests one edit at a time and can be steered in real time.

That's a critical feature for keeping a human in the loop, preventing big detours and token waste.

Most of these are Anthropic models under the hood, so I think 'whatever fits your workflow best' is the main deciding factor. That's definitely Claude Code for me, and I do think there's some 'secret sauce' in the exact prompting and looping logic they use, but I haven't tried Cursor a lot to be sure.
any secret sauce in prompting etc could be trivially reverse engineered by the companies building the other agents, since they could easily capture all the prompts it sends to the LLM. If there’s any edge, it’s probably more around them fine-tuning the model itself on Claude Code tasks.
Claude code seems like the obvious choice for someone using Vim but even in the context of someone using a graphical IDE like VSCode I keep hearing that Claude is “better” but I just can’t fathom how that can be the case.

Even if the prompting and looping logic is better, the direct integration with the graphical system along with the integrated terminal is a huge benefit, and with graphical IDEs the setup and learning curve is minimal.

If Cursor works for you, then stick with it. Claude Code is great for terminal-based workflows. Whatever makes you more productive is the better tool.

I’m just glad we’re getting past the insufferable “use Cursor or get left behind” attitude that was taking off a year ago.

Particularly with the VSCode extension. I was a loyal Cline user until recently (Windsurf/Cursor et. al. always felt grossly underpowered). But the ability for Claude to go off and run for 10+ minutes effectively autonomously is a gamechanger. The token usage has also gotten much more efficient in the last few months. With proper IDE support now I don't see any reason at all to use anything else, especially not the "credit" based middle-man providers.
if I don't see aider in the first sentence, I send it back