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

Claude Code: Best practices for agentic coding

anthropic.com

Post reply on HN
Print view View original

231–240 of 268 posts

Re: Claude Code: Best practices for agentic coding

#231

Post by gdudeman »

mr-karan wrote: post #180
simonw wrote: post #149

The "ultrathink" thing is pretty funny: > We recommend using the word "think" to trigger extended thinking mode, which gives Claude additional computation time to evaluate alternatives more thoroughly. These specific phrases are mapped directly to increasing levels of thinking budget in the system: "think" I had a poke around and it's not a feature of the Claude model, it's specific to Claude Code. There's a "megathi…

What I don’t like about Claude Code is why can’t they give command line flags for this stuff? It’s better documented and people don’t have to discover this the hard way. Similarly, I do miss an —add command line flag to manual specify the context (files) during the session. Right now I pretty much end up copy pasting the relative paths from VSCode and supply to Claude. Aider has much better semantics for such stuff.

Maybe I’m not getting this, but you can tab to autocomplete file paths.

You can use English or —add if you want to tell Claude to reference them.

Re: Claude Code: Best practices for agentic coding

#232

Post by flashgordon »

neodypsis wrote: post #191
flashgordon wrote: post #145

So I feel like a grandpa reading this. I gave Claude code a solid shot. Had some wins but costs started blowing up. I switched to Gemini AI where I only upload files I want it to work on and make sure to refactor often so modularity remains fairly high. It's an amazing experience. If this is any measure - I've been averaging about 5-6 "small features" per 10k tokens. And I totally suck at fe coding!! The other intere…

A combination that works nicely to solve bugs is: 1) have Gemini analyze the code and the problem, 2) ask it to create a prompt for Claude to fix the problem, 3) give Claude the markdown prompt and the code, 4) give Gemini the output from Claude to review, 5) repeat if necessary

Damn that's interesting. How much of the code do you provide? I'm guessing when modularity is high you can give specific files.

Re: Claude Code: Best practices for agentic coding

#233

Post by Wowfunhappy »

zomglings wrote: post #230
Wowfunhappy wrote: post #117

Earlier quoted context omitted.

An overview of the project and the features implemented. Edit: Here's the actual file if you want to see it. https://gist.github.com/Wowfunhappy/e7e178136c47c2589cfa7e5a...

Apologies for the late reply. My kids demanded my attention yesterday. It doesn't seem to have included any points on style or workflow in the context. Most of my context documents end up including the following information: 1. I want the agent to treat git commits as checkpoints so that we can revert really silly changes it makes. 2. I want it to keep on running build/tests on the code to be sure it isn't just going…

But none of this is really Claude Code's internal context, right? It's a summary. I could see using it as an alternative to /compact but not to undo a /clear.

Whatever the internal state is of Claude Code, it's lost as soon as you /clear or close the Terminal window. You can't even experiment with a different prompt and then--if you don't like the prompt--go back to the original conversation, because pressing esc to branch the conversation looses the original branch.

Re: Claude Code: Best practices for agentic coding

#234

Post by NiloCK »

simonw wrote: post #149

The "ultrathink" thing is pretty funny: > We recommend using the word "think" to trigger extended thinking mode, which gives Claude additional computation time to evaluate alternatives more thoroughly. These specific phrases are mapped directly to increasing levels of thinking budget in the system: "think" I had a poke around and it's not a feature of the Claude model, it's specific to Claude Code. There's a "megathi…

Slightly shameless, but easier than typing a longer reply.

https://www.paritybits.me/think-toggles-are-dumb/

https://nilock.github.io/autothink/

LLMs with broad contextual capabilities shouldn't need to be guided in this manor. Claude can tell a trivial task from a complex one just as easily as I can, and should self-adjust, up to thresholds of compute spending, etc.

Re: Claude Code: Best practices for agentic coding

#235

Post by jdiff »

oezi wrote: post #200
troupo wrote: post #132

Earlier quoted context omitted.

So, AIs are overeager junior developers at best, and not the magical programmer replacements they are advertised as.

As of April 2025. The pace is so fast that it will overtake seniors within years maybe months.

That's been said since at least 2021 (the release date for GitHub Copilot). I think you're overestimating the pace.

Re: Claude Code: Best practices for agentic coding

#236

Post by Terretta »

bob1029 wrote: post #210

I've developed a new mental model of the LLM codebase automation solutions. These are effectively identical to outsourcing your product to someone like Infosys. From an information theory perspective, you need to communicate approximately the same amount of things in either case. Tweaking claude.md files until the desired result is achieved is similar to a back and forth email chain with the contractor. The differenc…

We, mere humans, communicate our needs poorly, and undervisualize until we see concrete results. This is the state of us.

Faced with us as a client, the LLM has infinite patience at linear but marginal cost (relative to your thinking/design time cost, and the value of instant iteration as you realize what you meant to picture and say).

With offshoring, telling them they're getting it wrong is not just horrifically slow thanks to comms and comprehension latency, it makes you a problem client, until soon you'll find the do-over cost becomes neither linear nor marginal.

Don't sleep on the power of small fast iterations (not vibes, concrete iterations), with an LLM tool that commits as you go and can roll back both code and mental model when you're down a garden path.

Re: Claude Code: Best practices for agentic coding

#237

Post by neodypsis »

flashgordon wrote: post #232
neodypsis wrote: post #191

Earlier quoted context omitted.

A combination that works nicely to solve bugs is: 1) have Gemini analyze the code and the problem, 2) ask it to create a prompt for Claude to fix the problem, 3) give Claude the markdown prompt and the code, 4) give Gemini the output from Claude to review, 5) repeat if necessary

Damn that's interesting. How much of the code do you provide? I'm guessing when modularity is high you can give specific files.

Gemini's context is very long, so I can feed it full files. I do the same with Claude, but I may need to start from scratch various times, so Gemini serves as memory (and is also good that Gemini has almost no hallucinations, so it's great as a code reviewer for Claude's edits).

Re: Claude Code: Best practices for agentic coding

#238

Post by neodypsis »

vessenes wrote: post #197
neodypsis wrote: post #191

Earlier quoted context omitted.

A combination that works nicely to solve bugs is: 1) have Gemini analyze the code and the problem, 2) ask it to create a prompt for Claude to fix the problem, 3) give Claude the markdown prompt and the code, 4) give Gemini the output from Claude to review, 5) repeat if necessary

If you like this plan, you can do this from the command line: `aider --model gemini --architect --editor-model claude-3.7` and aider will take care of all the fiddly bits including git commits for you. right now `aider --model o3 --architect` has the highest rating on the Aider leaderboards, but it costs wayyy more than just --model gemini.

I like Gemini for "architect" roles, it has very good code recall (almost no hallucinations, or none lately), so it can successfully review code edits by Claude. I also find it useful to ground it with Google Search.

Re: Claude Code: Best practices for agentic coding

#239

Post by kkukshtel »

I recently wrote a big blog post on my experience spending about $200 with Claude Code to "vibecode" some major feature enhancements for my image gallery site mood.site

https://kylekukshtel.com/vibecoding-claude-code-cline-sonnet...

Would definitely recommend people reading it for some insight into hands on experience with the tool.

Re: Claude Code: Best practices for agentic coding

#240

Post by adastra22 »

simonw wrote: post #149

The "ultrathink" thing is pretty funny: > We recommend using the word "think" to trigger extended thinking mode, which gives Claude additional computation time to evaluate alternatives more thoroughly. These specific phrases are mapped directly to increasing levels of thinking budget in the system: "think" I had a poke around and it's not a feature of the Claude model, it's specific to Claude Code. There's a "megathi…

“think hard with a vengeance”
Post reply on HN