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

My experience with Claude Code after two weeks of adventures

sankalp.bearblog.dev

Post reply on HN
Print view View original

371–380 of 388 posts

Re: My experience with Claude Code after two weeks of adventures

#371

Post by swader999 »

oblio wrote: post #350
swader999 wrote: post #265

Earlier quoted context omitted.

I remember a few projects before all the AI where I would setup templates, little code generators, patterns to follow. Just make it really easy to do the right things. Was so easy to get things done. Was a small team, like minded, easy to row in the same direction. Your post reminded me of that. I too notice this about Claude, I've written commands, massaged Claude.md, even hooks. Given it very precise feature guides…

> My highest velocity was about 1.6 fib complexity points a day over thirty years, now it's 4.3 with Claude the last three weeks which is nuts. Where do you work that you managed to track what I assume are Agile story points over 30 years, also, so accurately? I don't even remember my median number for a sprint, let alone my daily average over long periods of time.

Yeah, accurately since 2008. Before that was just a guess but it wasn't at all fast. I've worked a lot of places but always insisted on story points based on estimated dev complexity.

Re: My experience with Claude Code after two weeks of adventures

#372

Post by fragmede »

actinium226 wrote: post #365
libraryofbabel wrote: post #58

Earlier quoted context omitted.

Sigh. As others have commented, over and over again in the last 6 months we've seen discussions on HN with the same basic variation of "Claude Code [or whatever] is amazing" with a reply along the lines of "It doesn't work for me, it just creates a bunch of slop in my codebase." I sympathize with both experiences and have had both. But I think we've reached the point where such posts (both positive and negative) are…

I'm not convinced that "we know they work really well for some people." So far I just see people really excited about the potential and really impressed at what it's capable of, but I think people are extrapolating poorly. It's like, yes it's impressive that it can make a video game with a few prompts, but that doesn't mean that with a few more prompts it'll turn into a AAA game. I'm on board with some limited AI aut…

If we handwave that the popular game Wordle, which made a lot of money for its author, could have been vibecoded, at what point does the gimmick become an actual feature that people look and pay for?

Re: My experience with Claude Code after two weeks of adventures

#373

Post by actinium226 »

fragmede wrote: post #372
actinium226 wrote: post #365

Earlier quoted context omitted.

I'm not convinced that "we know they work really well for some people." So far I just see people really excited about the potential and really impressed at what it's capable of, but I think people are extrapolating poorly. It's like, yes it's impressive that it can make a video game with a few prompts, but that doesn't mean that with a few more prompts it'll turn into a AAA game. I'm on board with some limited AI aut…

If we handwave that the popular game Wordle, which made a lot of money for its author, could have been vibecoded, at what point does the gimmick become an actual feature that people look and pay for?

No shade at wordle, but what you're describing sounds like it would be useful for the shovelware industry and that's about it. Not exactly a great leap forward for humanity...

Although I should be fair, this can help with one-off scripts that research folks usually do, when you just need to plot some data or do some back-of-the-terminal math. That said I don't think this would be a game changer, more of an efficiency boost and a limited one at that.

Re: My experience with Claude Code after two weeks of adventures

#374

Post by fragmede »

actinium226 wrote: post #373
fragmede wrote: post #372

Earlier quoted context omitted.

If we handwave that the popular game Wordle, which made a lot of money for its author, could have been vibecoded, at what point does the gimmick become an actual feature that people look and pay for?

No shade at wordle, but what you're describing sounds like it would be useful for the shovelware industry and that's about it. Not exactly a great leap forward for humanity... Although I should be fair, this can help with one-off scripts that research folks usually do, when you just need to plot some data or do some back-of-the-terminal math. That said I don't think this would be a game changer, more of an efficiency…

What would a great leap forwards for humanity look like? Sure, making it easier to shovel out shovelware means more shovelware, but why is that a bad thing? If customers have a very specific problem that wasn't going to get solved because it was too expensive to build a custom solution, and they now get to have bespoke software to cure their ills, other than being judgemental about this hypothetical piece of software as being shovelware, why is that a bad thing?

Re: My experience with Claude Code after two weeks of adventures

#375

Post by actinium226 »

fragmede wrote: post #374
actinium226 wrote: post #373

Earlier quoted context omitted.

No shade at wordle, but what you're describing sounds like it would be useful for the shovelware industry and that's about it. Not exactly a great leap forward for humanity... Although I should be fair, this can help with one-off scripts that research folks usually do, when you just need to plot some data or do some back-of-the-terminal math. That said I don't think this would be a game changer, more of an efficiency…

What would a great leap forwards for humanity look like? Sure, making it easier to shovel out shovelware means more shovelware, but why is that a bad thing? If customers have a very specific problem that wasn't going to get solved because it was too expensive to build a custom solution, and they now get to have bespoke software to cure their ills, other than being judgemental about this hypothetical piece of software…

Here's one version of what a great leap forward could look like, but it's simply one of many: an LLM that understand the CPU it's running on and can turn prompts into assembly, taking full advantage of the hardware. Or maybe it could target a virtual CPU like Java, but the point is that if the LLM can write code, why do it in Python or C? Just let it understand the CPU and let it rip. The only reason we have C/Python/etc. in the first place is because assembly sucks for humans to work with.

As to the shovelware, if it benefits people that's great, and I think the net benefit will likely be positive, but only slightly. The point in calling it shovelware is to suggest that it's low quality, and so it could have bugs and other performance issues that add costs to using which subtract from the benefit it provides (possibly in a net positive way, but probably not as fundamentally game changing as, say, Docker).

Re: My experience with Claude Code after two weeks of adventures

#376

Post by hamandcheese »

ModernMech wrote: post #347
hamandcheese wrote: post #168

For me the best part about AI is that when I'm feeling lazy, I can tell the AI to do it. Whether it gives me gold or gives me shit, it doesn't matter, because I have now started my work.

I thought this as well, but I just got burned in a way that caused me to lose so much time: I was feeling lazy and allowed AI to write some code for me. It looked good and compiled so I committed and pushed it. Weeks later I experienced a crash that I couldn't pinpoint. After hours of debugging, I eventually realized the culprit was that bit of code created by AI from weeks ago; it used a function that was defined in…

Was that being lazy with AI, or being lazy with code review? That doesn't strike me as any different than rubber stamping a colleagues code.

Re: My experience with Claude Code after two weeks of adventures

#377

Post by hamandcheese »

crinkly wrote: post #188
kaffekaka wrote: post #186

Earlier quoted context omitted.

Where in lies the dishonesty? Having the AI just getting the wheels turning when I'm not in the mood myself has many times been a good way to make progress.

When you say progress, I wonder at what cost. The apparatus for original thought isn't used and pushing through a problem and deriving the satisfaction of doing that doesn't happen. That leads to the wrong motivational cues. The brain is lazy and likes to take the easy route. Eventually you become merely a machine operator, which is something a lot of my colleagues are starting to do. And it's very worrying. The bias…

If you review some shit code and think of a better way to do it on the spot, is that original thought?

Does the answer change if the author is an LLM as opposed to another person?

Climb down from your high horse why don't you.

Re: My experience with Claude Code after two weeks of adventures

#378

Post by slashdev »

Wowfunhappy wrote: post #302
nico wrote: post #281

Earlier quoted context omitted.

> some of the individual changes it asks me to approve are too small for me to make a decision. There are cases where I almost denied a change initially, then realized Claude's approach made sense once I saw the full change set Yes, I've adapted to just review quickly, then if it makes sense as part of the task, let it keep going until it's done with the whole thing. Most of the times, by the end it does the right th…

> I love that it doesn't auto-commit everything, ala aider, so it's pretty painless to undo stuff. Yeah, I'm definitely glad it doesn't commit for me. The main issue I have is that I'm never sure how granular to make my commits. Sometimes I make them very granular because I'm experimenting with Claude and I want to be able to revert to any point in the conversation—but now I have to write a message each time to keep…

My commit messages have always been crap, so this doesn’t bother me much. I squash it all at the end anyway.

Still sometimes I forget to commit between tasks or it goes wild fixing compiler errors I didn’t want to address.

Re: My experience with Claude Code after two weeks of adventures

#379

Post by slashdev »

hamandcheese wrote: post #168

For me the best part about AI is that when I'm feeling lazy, I can tell the AI to do it. Whether it gives me gold or gives me shit, it doesn't matter, because I have now started my work.

Yes, it helps a lot overcome that initial momentum. I also find it helps at the end of the day when I’m tired. I can just tell it to do the work a little bit at a time and review/fix the output. Takes less out of me.

Re: My experience with Claude Code after two weeks of adventures

#380

Post by chisleu »

poszlem wrote: post #220
chisleu wrote: post #192

Earlier quoted context omitted.

Like working with an incredibly talented and knowledgable junior engineer, but still a junior engineer. If you want to try something better than claude code, try Cline.

Can you explain how cline is better?

I love the interface. It makes it extremely easy to rewind time to undo code edits and rewinding the LLM context at the same time. It's prompting and toolset is great. It's got MCP which I have integrated into my workflow. It's got a solid marketplace of auto installing MCP services. I love it.
Post reply on HN