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

311–320 of 388 posts

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

#311

Post by ak217 »

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…

The key to productivity with these tools (as with most things) is a tight feedback loop.

Cursor has an amazing tab completion model. It uses a bunch of heuristics to try to understand what you're doing and accelerate it. It's a bit like a very advanced macro engine, except I don't have to twist my brain into a pretzel trying to program my editor - it just works, and when it doesn't I press Esc, which takes about half a second. Or I reach for the agentic mode, where the iteration cycle is more like 30 seconds to a minute.

With the fully agentic editors, it takes more like 15 to 30 minutes, and now it's a full on interruption in my flow. Reviewing the agent's output is a chore. It takes so much more mental bandwidth than the quick accept/reject cycle in the editor. And I have to choose between giving it network access (huge security hazard) or keeping it offline (crippling it in most codebases).

I find that for throwaway codebases where I don't care about maintenance/security/reliability, I can put up with this. It's also incredibly useful when working with a language or framework that I don't fully understand (but the model understands better). But for most of my work it produces poor results and ends up being a net negative for productivity - for now.

I'm sure this will improve, but for now I consistently get better results in Cursor.

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

#312

Post by WorldMaker »

ActionHank wrote: post #274
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…

What people haven't realised yet is that Cursor isn't a product, it's a collection of features that every product is feverishly working to add. The key take away here is that save for the deep integration, there is a strategy for working with a best in class agent solution agnostic of tooling. These learnings will eventually coalesce into "best practices" that people will apply using the editor or IDE of their choice…

Arguably this is why I'm still long on GitHub Copilot. Love them or hate them Microsoft/GitHub are proven winners in feature parity work and have such a big moat today. Why use a VS Code fork when you can "use the real thing"?

Microsoft/GitHub has even signaled confidence that they can compete feature by feature with the forks by even open sourcing much of the Copilot integration in VS Code.

Aesthetically, I like VS Code Copilot/Copilot Chat UIs/UX for the most part, certainly better than I like Claude Code and more than I like Cursor, too.

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

#313

Post by FiberBundle »

TheRoque wrote: post #88
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.

100% agree, I have been looking for a YouTube video or stream of someone leveraging AI to get their productivity boost, but I haven't found anything that made me think "okay, that really speeds up things"

https://m.youtube.com/watch?v=XyQ4ZTS5dGw&pp=ygUZTWl0Y2hlbGw...

Not a "100x" boost, but a pretty good take on what tasks agents can do for even very good programmers.

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

#314

Post by chis »

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…

So far my feeling is that cursor is a bit better for working on existing large codebases, whereas claude does a good job planning and building greenfield projects that require a lot of code.

That does net out to meaning that Cursor gets used way more often, atm.

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

#315

Post by crinkly »

malthaus wrote: post #262
crinkly wrote: post #237

Earlier quoted context omitted.

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

There's more than just me. We keep each other honest.

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

#316

Post by ookblah »

filoeleven wrote: post #264
ookblah wrote: post #121

Earlier quoted context omitted.

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.

i mean at this point i don't really know what to say to comments like these basically calling me a liar lol. i log increases in sustained lines outputted and removed, files touched, features pushed per day. you can argue the level of increase, but it's definitely not 20% slower shrugs.

and this is exactly what i'm talking about. in the end who gives a shit about some study that may or may not apply to me as long its actually working.

people literally frothing at the mouth to tell people who find it useful that they are utterly wrong or delusional. if you don't like it then just go about your day, thx.

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

#317

Post by varispeed »

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…

Looks like I am not the one. I look at the Claude Code reviews and feel as if I had a stroke. What is it that I am not getting?

The whole thing looks like one large source of friction.

I've been using Cursor a lot and it speeds up everything for me. Not sure what kind of benefit Claude Code would give me.

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

#318

Post by steveklabnik »

dude250711 wrote: post #207
nikcub wrote: post #172

Earlier quoted context omitted.

because their philosophy is that developers are tied to their editors and IDE's so you should be able to bring you're own. and they're right.

Where are the awesome deeply-integrated add-ins/extensions for all popular IDEs then?

Right now, they have ones for VS: Code and JetBrains' various ones: https://docs.anthropic.com/en/docs/claude-code/ide-integrati...

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

#319

Post by varispeed »

nico wrote: post #243
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 used to feel similarly. But recently started using Claude Code and it does feel a lot better than Cursor for me I'm not sure why. However, Claude does seem to know better where things are and knows not to make unnecessary changes. I still need to guide it and tell it to do things differently some times, but it feels like it's a lot more effective Personally, I also like that usually it just presents to me only one…

I rarely had Cursor do unwanted changes for me. Maybe this is about prompting? I am very particular with what I want and I try to be as verbose as I can and explain the context as well plus hint which files I believe it should put attention to.

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

#320

Post by extr »

storus wrote: post #227
danielbln wrote: post #209

Earlier quoted context omitted.

It's the white page problem, that LLMs solve handily. Instead of facing the daunting task of building back up a complex mental model, I can ask the machine "yo, what were we doing? what's this thing? ok, how about you give it a spin and show me" and all of the sudden I'm back in it and can get cracking. That, and rubber ducking, and speeding up yak shaving to the nth degree really make this thing very useful to me. I…

Maybe there is a reason why is your brain hesitating (e.g. low energy) and pushing it with LLMs would blow it up at some point in the future if the cause was not fixed in the meantime? Covid showed me (in a very accentuated way) that many of the times I was procrastinating was instead brain throttling due to low energy, protecting itself.

I think for me the hesitation is in that to get started on task XYZ, the first step is often some extremely boring task ABC that is only tangentially related to the "real" task.

For example, the other day I needed to find which hook to use in a 2300 loc file of hooks. AI found me the hook in 5 seconds and showed me how to use it. This is a pure win - now that I have the name of the hook I can go read it in 30 seconds and verify it's what I wanted. If it's not, I can ask again. There's zero risk here.

Post reply on HN