Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

6 weeks of Claude Code

blog.puzzmo.com

Post reply on HN
Print view View original

371–380 of 603 posts

Re: 6 weeks of Claude Code

#371

Post by distances »

dvfjsdhgfv wrote: post #207
msikora wrote: post #159

Earlier quoted context omitted.

Just a few months ago I couldn't imagine paying more than $20/mo for any kind of subscription, but here I am paying $200/mo for the Max 20 plan! Similarly amazed as an experienced dev with 20 YoE (and a fellow Slovak, although US based). The other tools, while helpful, were just not "there" and they were often simply more trouble than they were worth producing a lot of useless garbage. Claude Code is clearly on anoth…

May I ask what you are use it for? I have been using it for fun mostly, side projects, learning, experimenting. I would never use it for work codebase, unless, well, the company ordered or at least permitted it. And even then, I'm not really sure I would feel comfortable with the level of liberty CC takes. So I'm curious about others.

Of course you need an explicit permit from the company to use (non-local) AI tools.

Before that was given, I used AI as a fancier search engine, and for coming up with solutions to problems I explained in abstract (without copy-pasting actual code in or out).

Re: 6 weeks of Claude Code

#372

Post by boredhedgehog »

philbo wrote: post #362

As one of the curious minority who keeps trying agentic coding but not liking it, I've been looking for explanations why my experience differs from the mainstream. I think it might lie in this nugget: > I believe with Claude Code, we are at the > “introduction of photography” period of > programming. Painting by hand just doesn’t > have the same appeal anymore when a single > concept can just appear and you shape it…

> still people paint, still people pay for paintings

But in what environment? It seems to me that most of the crafts that have been replaced by the assembly line are practiced not so much for the product itself, but for an experience both the creator and the consumer can participate in, at least in their imagination.

You don't just order such artifacts on Amazon anonymously; you establish some sort of relationship with the artisan and his creative process. You become part of a narrative. Coding is going to need something similar if it wants to live in that niche.

Re: 6 weeks of Claude Code

#373

Post by EFreethought »

illusive4080 wrote: post #319
EFreethought wrote: post #318

Earlier quoted context omitted.

> It can, in fact, control your entire computer. Honestly, that sounds like malware.

No because _you_ are in control. Not the AI.

Maybe you do today. Will that always be the case? People running Windows do not have as much control over their systems as they should. What does enshittification look like for AI?

Why should I trust any of these AI vendors?

Re: 6 weeks of Claude Code

#374

Post by AtNightWeCode »

The problem for me is to predict what AI might get wrong. Claude can solve hard coding problems one day just to fail with basic stuff like thread safety the next. But overall I think it is clear that we reached the point where AI, if used correctly, saves a lot of development time.

Re: 6 weeks of Claude Code

#375

Post by lukaslalinsky »

bsder wrote: post #365
lukaslalinsky wrote: post #57

I have about two weeks of using Claude Code and to be honest, as a vibe coding skeptic, I was amazed. It has a learning curve. You need to learn how to give it proper context, how to chunk up the work, etc. And you need to know how to program, obviously. Asking it to do something you don't know how to do, that's just asking for a disaster. I have more than 25 years of experience, so I'm confident with anything Claude…

> I have about two weeks of using Claude Code and to be honest, as a vibe coding skeptic, I was amazed. And, yet, when I asked it to correct a CMake error in a fully open source codebase (broken dependency declaration), it couldn't work it out. It even started hallucinating version numbers and dependencies that were so obviously broken that at least it was obvious to me that it wasn't helping. This has been, and cont…

This is typically the outcome, when you have it look at a generic problem and fix it, especially if the problem depends on external information (like specific version numbers, etc). You have to either tell it where to look it up, or ask it to ask you questions how things need to be resolved. I personally use it to work on native code, C++ (with CMake), Zig, some Python. Works fine.

Re: 6 weeks of Claude Code

#376

Post by __loam »

stitched2gethr wrote: post #263
einpoklum wrote: post #150

Earlier quoted context omitted.

It's not a dream come true to have a bunch of GPUs crunching at full power to achieve your minor automation, with the company making them available losing massive amounts of money on it: https://www.wheresyoured.at/the-haters-gui/ ... while also exposing the contents of your computer to surveillence.

Well me plus a $1200 / year subscription is still much cheaper than 2 of me.

The point is that it costs more than $1200, you're just not the one paying all the costs. It seems like there are a ton of people on HN who are absolutely pumped to be totally dependent on a tool that must rugpull them eventually to continue existing as a business. It feels like an incredible shame that the craft is now starting to become totally dependent on tools like this, where you're calling out to the cloud to do even the most basic programming task.

Re: 6 weeks of Claude Code

#377

Post by hansvm »

AndyNemmity wrote: post #353
hansvm wrote: post #351

Earlier quoted context omitted.

I don't think that's true with current-gen models. You can even go so far as to write pseudocode for the LLM to translate to a real language, and for anything out-of-the-box my experience is that it will blatantly ignore your instructions. A baseline-competent junior at least has the context to know that if there are 5 steps listed and they only did 3 then there's probably a problem. Prompting an LLM is definitely a…

I don't believe it is good enough, it's also not as relevant. my prompts have gotten less and less. It's the hooks and subagents, and using the tools that matter far more. This is a thread about claude code. the other LLms don't matter. Nothing ever blatantly ignores my instructions in claude code. that's a thing of the past. Of course, not using claude code, for sure. But all my instructions are followed with my set…

Your experience echoes my own for sufficiently trivial tasks, but I haven't gotten any of this to work for the actual time-consuming parts of my job. It's so reliably bad for some tasks that I've reworked them into screening questions for candidates trying to skate by with AI without knowing the fundamentals. Is that really not your experience, even with claude code?

Re: 6 weeks of Claude Code

#378

Post by troupo »

Aurornis wrote: post #77
lukaslalinsky wrote: post #57

I have about two weeks of using Claude Code and to be honest, as a vibe coding skeptic, I was amazed. It has a learning curve. You need to learn how to give it proper context, how to chunk up the work, etc. And you need to know how to program, obviously. Asking it to do something you don't know how to do, that's just asking for a disaster. I have more than 25 years of experience, so I'm confident with anything Claude…

> as a vibe coding skeptic, I was amazed. The interesting thing about all of this vibe coding skepticism, cynicism, and backlash is that many people have their expectations set extremely low. They’re convinced everything the tools produce will be junk or that the worst case examples people provide are representative of the average. Then they finally go out and use the tools and realize that they exceed their (extreme…

> The interesting thing about all of this vibe coding skepticism, cynicism, and backlash is that many people have their expectations set extremely low.

Or they have actually used all these tools, know how they work, and don't buy into hype and marketing.

Re: 6 weeks of Claude Code

#379

Post by AndyNemmity »

hansvm wrote: post #377
AndyNemmity wrote: post #353

Earlier quoted context omitted.

I don't believe it is good enough, it's also not as relevant. my prompts have gotten less and less. It's the hooks and subagents, and using the tools that matter far more. This is a thread about claude code. the other LLms don't matter. Nothing ever blatantly ignores my instructions in claude code. that's a thing of the past. Of course, not using claude code, for sure. But all my instructions are followed with my set…

Your experience echoes my own for sufficiently trivial tasks, but I haven't gotten any of this to work for the actual time-consuming parts of my job. It's so reliably bad for some tasks that I've reworked them into screening questions for candidates trying to skate by with AI without knowing the fundamentals. Is that really not your experience, even with claude code?

Right, and I wasn't able to get this to work for any actual time consuming parts of my job until last weekend with sub-agents, and testing head to head battles with sub-agents, and selecting the best one and repeating.

Last weekend I did nothing but have different ideas battle it out against each other, with me selecting the most successful one, and repeating.

And now, my experience is no longer the same. Before last weekend, i had the same experience you are describing.

Re: 6 weeks of Claude Code

#380

Post by globular-toast »

jaggederest wrote: post #255
globular-toast wrote: post #212

Earlier quoted context omitted.

Just show the code you've produced.

Right but that's like showing the compiled binary but not the source. With claude code prompt and config is everything.

This is the saddest part of this whole thing for me. You consider the prompts and config to be the real source code, but those are just completely lost into the ether. Even if you saved the prompts you can't reproduce their effects.

Then there's the question of how do other developers contribute to the code. They don't have your prompts, they just have the code.

So, no, prompts are not source code, that's why I ask for people to just show the code they are producing and nobody ever does.

Post reply on HN