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

261–270 of 388 posts

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

#261

Post by swader999 »

muzani wrote: post #240
ianberdin wrote: post #232

Reading all these glowing reviews of Claude Code, I still get the feeling that either everyone’s been paid off or it’s just the die-hard fans of terminal windows and editors like Emacs and Vim. Using the terminal is right up their alley—it’s in their DNA. Every time I read comments saying Claude Code is far better than Cursor, I fire it up, pay for a subscription, and run it on a large, complex TypeScript codebase. F…

From what I see, you have to get your claude.md instructions right, and most of it is using planning mode. The CLI is exactly how you interact with it. It's less vibe code and what I'd call supervisory code. It's less about writing code, more about building an engine that codes. It's hands off but unlike Cursor, it's not as eyes-off. You do observe and interrupt it. It is great for full on TDF - it figures out the fe…

I'm just full of joy building again. It's hard being a manager, kids, jaded with the front end framework dance. Now I can just build build build and get my ideas out to users faster than they can decide.

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

#262

Post by malthaus »

crinkly wrote: post #237
scrollaway wrote: post #233

Earlier quoted context omitted.

People most likely pay you to do valuable work, not quality work. One doesn’t necessarily mean the other. If your coworkers can get into the groove faster than you (for whatever reason), you’ll continue to do quality work, not look busy, and no longer be paid for it.

Quality is the value here. Because low quality has a financial risk. And good luck finding anyone these days who can do quality. It's a nice niche.

it sounds like they're buying your confidence primarily

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

#263

Post by artisin »

ants_everywhere wrote: post #89
_se wrote: post #44

Earlier quoted context omitted.

It's always POC apps in js or python, or very small libraries in other popular languages with good structure from the start. There are ways to make them somewhat better in other cases (automated testing/validation/linting being a big one), but for the type of thing that 95% of developers are doing day to day (working on a big, sprawling code base where none of those attributes apply), it's not close to being there. T…

> They're amazing. But the moment you try to do the more "serious" work with them, it falls apart rapidly. Sorry, but this is just not true. I'm using agents with a totally idiosyncratic code base of Haskell + Bazel + Flutter. It's a stack that is so quirky and niche that even Google hasn't been able to make it work well despite all their developer talent and years of SWEs pushing for things like Haskell support inte…

Hold the phone. So, Google, with its legions of summa cum laude engineers, can't make this stack work well, but your AI agent is nailing it into next week? Seriously, show me the way, so I too may find AI enlightenment.

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

#264

Post by filoeleven »

ookblah wrote: post #121
erentz wrote: post #50

There must at this point be lots and lots of actual walkthroughs of people coding using Claude Code, or whatever, and producing real world apps or libraries with them right? Would be neat to have a list because this is what I want to read (or watch), rather than people just continuously telling me all this is amazing but not showing me it’s amazing.

this sounds like a cop out, but honestly you will probably see the most vocal on both sides of this here while vast majority are just quietly doing working and doing their stuff (ironic i'm writing this). i feel like some kind of shill, but honestly i'm anywhere from 1.5x to 10x on certain tasks. the main benefit is that i can reduce a lot of cognitive load on tasks where they are either 1) exploratory 2) throwaway 3…

> i'm anywhere from 1.5x to 10x on certain tasks

By your own reckoning. There was that recent study showing how senior devs using AI thought they were 20% faster, when they were objectively 20% slower.

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

#265

Post by swader999 »

ianberdin wrote: post #244
ianberdin wrote: post #232

Reading all these glowing reviews of Claude Code, I still get the feeling that either everyone’s been paid off or it’s just the die-hard fans of terminal windows and editors like Emacs and Vim. Using the terminal is right up their alley—it’s in their DNA. Every time I read comments saying Claude Code is far better than Cursor, I fire it up, pay for a subscription, and run it on a large, complex TypeScript codebase. F…

And so you don’t think I just dropped in to try it out: no, I’ve been using Cursor every day for many months. I set up the rules very precisely, make maps, write instructions and constraints so the LLM understands how to work with the current codebase—where to find things, what to do, what not to do, what things should look like. I try to write as concisely as possible so everything fits in the context window. I have…

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. I see the same issues you do. Feels like Claude has a lazy lever built in to it.

I play GPT O3 off it constantly, it takes the spec, a repomix slice of the Claude work and gives a new gap.md for Claude to pursue. Rinse and repeat. It works but your cursor flow seems better.

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. I'm a total hack, I think I could get it up to nine if I was a bit more organized. Probably won't have to, just wait for the next iteration of the Claude models.

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

#266

Post by swader999 »

My biggest problem at the moment is that I am more inclined to talk to Claude and GPT O3 than the other more junior devs on my team. I just get more done. I'm not being facetious here, I really don't like that about the way I've been the past few days. I have to use the new free time I have to work more with them I guess.

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

#267

Post by Waterluvian »

ianberdin wrote: post #232

Reading all these glowing reviews of Claude Code, I still get the feeling that either everyone’s been paid off or it’s just the die-hard fans of terminal windows and editors like Emacs and Vim. Using the terminal is right up their alley—it’s in their DNA. Every time I read comments saying Claude Code is far better than Cursor, I fire it up, pay for a subscription, and run it on a large, complex TypeScript codebase. F…

I think it’s because a lot of people deeply misunderstand what the hard parts of software development actually are. Most of the time, programs won’t involve novel or complicated algorithms or whatnot. It’s mostly just gluing ideas together. But that all comes after the specification, design, architecture, planning, etc.

I think it can also be deceptive the same way it can be in basically any trade: it’s easy for the job to get “done” and work. It’s hard to do it properly in a way that it has a lasting durability. With A.I. we can spit out these programs that work and look done. And for prototypes or throwaways and such, that’s probably wonderful! But that’s not going to fly if you’re building a house to spec for people who are going to live in it for 30 years.

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

#268

Post by two_tasty »

intrasight wrote: post #239
darkwater wrote: post #195

Earlier quoted context omitted.

Do tests let you understand a codebase you have not written?

this desire to understand code will be soon be seen as rather anachronistic. What's important is that you understand your tests. Let the AI generate the code. The spec and the test are your human contribution.

Ah yes, can't wait to tell my auditor / regulator "I don't understand the code because Claude wrote it, but it's fine, because understand the code is for boomers." That will get a big laugh in a deposition.

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

#269

Post by kierangill »

Past the basic setup, what are people doing to grease the tracks for these tools?

That is, how are people organizing their context and codebase to help the tooling guide themselves to the right answer?

I have some half-baked thoughts here [1], but I know there are more (better) methodologies to be discovered.

[1]: https://blog.kierangill.xyz/oversight-and-guidance

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

#270

Post by jmkni »

swader999 wrote: post #266

My biggest problem at the moment is that I am more inclined to talk to Claude and GPT O3 than the other more junior devs on my team. I just get more done. I'm not being facetious here, I really don't like that about the way I've been the past few days. I have to use the new free time I have to work more with them I guess.

I guess the inverse is true too though, junior devs are more inclined to ask AI than their superiors.
Post reply on HN