stitched2gethr

Without saying that I think LLMs are alive, I do think it matters. I have personally been cruel to an LLM in ways that would make me ashamed if I suddenly understand that it had feelings.

Try this: `claude --model claude-opus-4-6`

> Anthropic already has a tool for it Yes, but Anthropic built this tool after OpenClaw, because of OpenClaw.

A question asking how your project compares to x so I don't have to do the research for myself.

This is intriguing.

On another note, I do not understand how posts make it to the top of the front page with essentially no comments.

Try proxymock. It's not open source but it is free to use.

It's not ready to make important decisions. But that's not the same as making important contributions.

We're way past that.

Insane. I'm terrified and I can't wait.

It all depends on the model and how much you use it of course. We're running Opus 4.6 and on a light day it spends a dollar or two. This is just a few simple operations like "create a ticket for ..." and it's regular heartbeat checks. The heaviest day I see is $110 and on that day we were basically talking to it and having it implement features all day long.

We did the same and I wrote (admittedly had AI write) about it.

https://speedscale.com/blog/building-speedy-autonomous-ai-de...

I had to scroll too far to find this take. 100%.

This is like saying the CLAUDE.md or AGENTS.md is irrelevant because the LLM generated it.

It's all about how full the context is, right? For a task that can be completed in 20% of the context it doesn't matter, but you don't want to fill your context with exploration before you do the hard part.

I have actually found something close to the opposite. I work on a large codebase and I often use the LLM to generate artifacts before performing the task (for complex tasks). I use a prompt to say "go explore this area if the code and write about it". It documents concepts and has pointers to specific code. Then a fresh session can use that without reading the stuff that doesn't matter. It uses more tokens overall, but includes important details that can get totally missed when you just let it go.

I think that's partly the point. This is the tool that everyone wanted but couldn't quite describe. Not saying he's a genius, but he was the first to will it into existence.

Try running `/insights` with Claude Code.

I think the rationale is that with the right tools you can move much faster, and not burn everything to the ground, than just rawdogging Claude. If you haven't bothered setting up extra tools you may still be faster / better than old you, but not better than the you that could be. I'm not preaching, that's just the idea.

> That is not software engineering or development, it's brogrammer trash.

Yes, but it's working. I'm still reading the code and calling out specific issues to Claude, but it's less and less.

I think it also presumes that the skills of today won't be helpful in making you better, faster, stronger at knowing what to learn tomorrow. Skateboarding ain't snowboarding but I guarantee the experience helps.

And how are you using it now? Have you seen real value weeks or months on?

This is what we're working on at Speedscale. Our methods use traffic capture and replay to validate what worked before still works today.

Perhaps if he was able to get Claude Code to do what he wanted in less time, and with a better experience, then maybe that's not a skill he (or the rest of us) want to develop.

Getting an old codebase working is something they are especially good at because it has regular, actionable feedback and it's clear when the tall is complete. Creating the "best" anything is much more open ended.

I'm surprised there was no mention of shorts which take attention away from long form content. It's all they a pushing these days. In my feed I regularly see 6 shorts, then a video, then 6 more shorts. TikTok is not the ideal model for everyone.

I've seen it many times. And then every task takes longer than the last one, which is what pushes teams to start rewrites. "There's never enough time to do it right, but always time to do it again."

Na. Consenting adults and all that.

There were developed around the same time so maybe that accounts for some of the comparisons, but at least in this case I think it matters that they are both relatively new languages with modern tooling.

This contains some specific data with pretty graphs: https://youtu.be/tbDDYKRFjhk?t=623

But if you do professional development and use something like Claude Code (the current standard, IMO) you'll quickly get a handle on what it's good at and what it isn't. I think it took me about 3-4 weeks of working with it at an overall 0x gain to realize what it's going to help me with and what it will make take longer.

> I think you've actually lost that time in the first case. And in the second case, as opposed to the middle ground.