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

131–140 of 388 posts

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

#131

Post by aniforprez »

johnfn wrote: post #105
whatever1 wrote: post #6

Earlier quoted context omitted.

I find amazing all the effort that people put trying to program a non deterministic black box. True courage.

You might also find it amazing that people work with colleagues and give them feedback!

Colleagues take that feedback and improve. It seems a large number of people here seem to find these kind of improvements in other humans useless in the hypothetical case of them leaving but I occasionally look at my mentee's LinkedIn and feel a sense of pride at having contributed. I also feel joy when they appreciate my efforts.

I genuinely don't understand how often people compare AI to junior developers.

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

#132

Post by iambateman »

pragmatic wrote: post #13
iambateman wrote: post #7

Claude Code is hard to describe. It’s almost like I changed jobs when I started using it. I’ve been all-in with Claude as a workflow tool, but this is literally steroids. If you haven’t tried it, I can’t recommend it enough. It’s the first time it really does feel like working with a junior engineer to me.

Could you elaborate a bit on the tasks,languages,domain etc you’re using it with? People have such widely varying experiences and I’m wondering why.

I’m a TALL developer, so Laravel, Livewire, Tailwind, Alpine.

It’s nice because 3/4 of those are well-known but not “default” industry choices and it still handles them very well.

So there’s a Laravel CRM builder called Filament which is really fun to work in. Claude does a great job with that. It’s a tremendous amount of boilerplate with clear documentation, so it makes sense that Claude would do well.

The thing I appreciate though is that CC as an agent is able to do a lot in one go.

I’ve also hooked CC up to a read-only API for a client, and I need to consume all of the data on that API for the transition to a Filament app. Claude is currently determining the schema, replicating it in Laravel, and doing a full pull of API records into Laravel models, all on its own. It’s been running for 10 minutes with no interruption and I expect will perform flawlessly at that.

I invest a lot of energy in prompt preparation. My prompts are usually about 200 words for a feature, and I’ll go back and forth with an LLM to make sure it thinks it’s clear enough.

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

#133

Post by baka367 »

arealaccount wrote: post #12
iambateman wrote: post #7

Claude Code is hard to describe. It’s almost like I changed jobs when I started using it. I’ve been all-in with Claude as a workflow tool, but this is literally steroids. If you haven’t tried it, I can’t recommend it enough. It’s the first time it really does feel like working with a junior engineer to me.

Weirdly enough I have the opposite experience where it will take several minutes to do something, then I go in and debug for a while because the app has become fubar, then finally realize it did the whole thing incorrectly and throw it all away. And I reach for Claude quite a bit because if it worked as well for me like everyone here says, that would be amazing. But at best it’ll get a bunch of boilerplate done after…

For me it fixed a library compatibility issue with React 19 in 10 mins and several nudges startign from the console error and library name.

It would have been a half-day worth of adventure at least should i have done it myself (from diagnosing to fixing)

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

#134

Post by komali2 »

One thing I'm slightly anxious about in this new LLM world is whether the prices I'm paying are sustainable. I crank the fuck out of cursor and I think we're paying like 40 bucks a month for the privilege. Is this early Uber where it was unbelievable how cheap the rides were? In 2030 am I going to have gotten dependent on ai assisted levels of productivity, be making client and customer promises based on that expectation, but suddenly find myself looking at 1k+ bills, now that all the ai companies need to actually make money?

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

#135

Post by komali2 »

iambateman wrote: post #7

Claude Code is hard to describe. It’s almost like I changed jobs when I started using it. I’ve been all-in with Claude as a workflow tool, but this is literally steroids. If you haven’t tried it, I can’t recommend it enough. It’s the first time it really does feel like working with a junior engineer to me.

Does anyone have any usage guides they can recommend to feel this way about using Claude code, other than the OP article? I fired it up yesterday for about an hour and tried it on a couple tickets and it felt like a total waste of time. The answers it gave were absurdly incorrect - I was being quite specific in my prompting and it seemed to be acquiring the proper context, but just doing nothing like what I was asking.

E.g. I asked it to swap all on change handlers in a component to modify a use State rather than directly fire a network request, and then add on blurs for the actual network request. It didn't add use states and just added on blurs that sent network requests to the wrong endpoint. Bizarre.

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

#136

Post by bugglebeetle »

gonzo41 wrote: post #112
taude wrote: post #51

Earlier quoted context omitted.

to continue on this, I wouldn't let claude or any agent actually create a project structure, i'd guide it in the custom system prompt. and then in each of the folders continue to have specific prompts for what you expect the assets to be coded like, and common behavior, libraries, etc....

So you've invented writing out a full business logic spec again. btw, I'm not throwing shade. I personally think upfront design through a large lumbering document is actually a good way to develop stuff. As you either do it upfront, or through endless iterations in sprints for years.

Yeah, my experience of working with Claude Code is that I’m actually far more conscientious about design. After using it for awhile, you get a good sense of its limits and how you need to break things down and spell thing out to overcome.

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

#137

Post by dmix »

yoyohello13 wrote: post #122
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.

Something really feels off about the whole thing. I use Claude code. I like it, it definitely saves me time reading docs or looking on stack overflow. It’s a phenomenal tool. If we are to believe the hype though, shouldn’t these tools be launching software into the stratosphere? Like the CEO of stripe said AI tools provide a x100 increase in productivity. That was 3-4months ago. Shouldn’t stripe be launching rockets…

Whatever it produces still needs to be carefully reviewed and guided. Context switching as a human programmer is very hard so you need to focus on the same specific task, which is harder to not switch to social media or IRL while waiting for it. And you're going to be on the same branch for the same ticket, git doesn't let you do multiple at once (at least I'm not set up to). Not sure where the productivity scaling would come from outside of rapid experimentation on bad ideas until you find the right one, and of course rapid autocomplete, faster debugging, much fancy global find/replace type stuff.

I use it quite aggressively and I'd probably only estimate 1.5x on average.

Not world changing because we all mostly work on boring stuff and have endless backlogs.

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

#138

Post by makeramen »

dawnerd wrote: post #80
taude wrote: post #42

Earlier quoted context omitted.

do you create the claude.md files at several levels of your folder structure, so you can teach it how to do different things? Configuring these default system prompts is required to get it to work well. I'd definitely watch Boris's intro video below [1] [1] Boris introduction: https://www.youtube.com/watch?v=6eBSHbLKuN0 [2] summary of above video: https://www.nibzard.com/claude-code/

By the time you do all of that you might as well just write code by hand.

You don't do it manually. You have claude do it once you’ve guided it back on track to remind itself not to do it next time.

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

#139

Post by satisfice »

iambateman wrote: post #7

Claude Code is hard to describe. It’s almost like I changed jobs when I started using it. I’ve been all-in with Claude as a workflow tool, but this is literally steroids. If you haven’t tried it, I can’t recommend it enough. It’s the first time it really does feel like working with a junior engineer to me.

Are you doing anything useful? How can anyone outside of yourself know this?

My own experiments only show that this technology is unreliable.

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

#140

Post by roywiggins »

ctoth wrote: post #37
blub wrote: post #29

Earlier quoted context omitted.

The contractors working on my house sometimes paint a room bright pink for no particular reason. When I point that out, they profusely apologize and say that of course the walls must be white and wonder why they even got the idea of making them pink in the first place. Odd, but nice fellows otherwise. It feels like they’re 10x more productive than other contractors.

I asked my contractor to install a door over the back stairs opening outward, came back, and it was installed opening inward. He told me he tried to figure out a way he could do it in-code, but there wasn't one, so that's what he had to do. I was slightly miffed he didn't consult me first, but he did the pragmatic thing. This actually happened to me Monday. But sure, humans are deterministic clockwork mechanisms! Are…

The big difference is that when questioned he didn't profusely apologize and immediately get to work undoing it, but instead gave you a reason that was probably causally related to the decision he made and not a retroactive justification made up on the spot.
Post reply on HN