Claude Code: Best practices for agentic coding

https://www.anthropic.com/engineering/claude-code-best-practices

by sqs • 1 year ago

614 250 1 year ago

The most interesting part of this article for me was:

Have multiple checkouts of your repo

I don’t know why this never occurred to me probably because it feels wrong to have multiple checkouts, but it makes sense so that you can keep each AI instance running at full speed. While LLM‘s are fast, this is one of the annoying parts of just waiting for an instance of Aider or Claude Code to finish something.

Also, I had never heard of git worktrees, that’s pretty interesting as well and seems like a good way to accomplish effectively having multiple checkouts.

m0rde• 1 year ago

I've never used Claude Code or other CLI-based agents. I use Cursor a lot to pair program, letting the AI do the majority of the work but actively guiding.

How do you keep tabs on multiple agents doing multiple things in a codebase? Is the end deliverable there a bunch of MRs to review later? Or is it a more YOLO approach of trusting the agents to write the code and deploy with no human in the loop?

rfoo• 1 year ago

In the same way how you manage a group of brilliant interns.

oxidant3• 1 year ago

Multiple terminal sessions. Well written prompts and CLAUDE.md files.

I like to start by describing the problem and having it do research into what it should do, writing to a markdown file, then get it to implement the changes. You can keep tabs on a few different tasks at a time and you don't need to approve Yolo mode for writes, to keep the cost down and the model going wild.

cadamsdotcom• 1 year ago

You might want to consider Claude Squad, https://github.com/smtg-ai/claude-squad which manages all the worktrees for you.

Disclaimer, I haven’t tried it personally - if you do, let us know how you go!

remoquete• 1 year ago

What's the Gemini equivalent of Claude Code and OpenAI's Codex? I've found projects like reugn/gemini-cli, but Gemini Code Assist seems limited to VS Code?

peterldowns2• 1 year ago

I would also like to know — I think people are using Cursor/Windsurf/Roo(Cline) for IDEs that let you pick the model, but I don't know of a CLI agentic editor that lets you use arbitrary models.

manojlds• 1 year ago

https://aider.chat/

jasir• 1 year ago

There's Aider, Plandex and Goose, all of which let you chose various providers and models. Aider also has a well known benchmark[0] that you can check out to help select models.

- Aider - https://aider.chat/ | https://github.com/Aider-AI/aider

- Plandex - https://plandex.ai/ | https://github.com/plandex-ai/plandex

- Goose - https://block.github.io/goose/ | https://github.com/block/goose

[0] https://aider.chat/docs/leaderboards/

boredtofears2• 1 year ago

I've only user aider (which I like quite a bit more than cursor) but I'm curious how it compares to plandex and goose.

jhawk28• 1 year ago

Junie from Jetbrains was recently released. Not sure what LLM is uses.

nojs• 1 year ago

Claude

zomglings13• 1 year ago

If anyone from Anthropic is reading this, your billing for Claude Code is hostile to your users.

Why doesn’t Claude Code usage count against the same plan that usage of Claude.ai and Claude Desktop are billed against?

I upgraded to the $200/month plan because I really like Claude Code but then was so annoyed to find that this upgrade didn’t even apply to my usage of Claude Code.

So now I’m not using Claude Code so much.

fcoury5• 1 year ago

I totally agree with this, I would rather have some kind of prediction than using the Claude Code roulette. I would definitely upgrade my plan if I got Claude Code usage included.

datavirtue3• 1 year ago

I don't what you guys are on about but I have been using the free GitHub Copilot in VS Code chats to absolutely crank out new UI features in Vue. All that stuff that makes you groan at the thought of it: more divs, bindings, form validation, a whole new widget...churned out in 30 seconds. Try it live. Works? Keep.

I'm surprised at the complexity and correctness at which it infers from very simple, almost inadequate, prompts.

zomglings13• 1 year ago

They did it!

cypherpunks012• 1 year ago

Claude Pro and other website/desktop subscription plans are subject to usage limits that would make it very difficult to use for Claude Code.

Claude Code uses the API interface and API pricing, and writes and edits code directly on your machine, this is a level past simply interacting with a separate chat bot. It seems a little disingenuous to say it's "hostile" to users, when the reality is yeah, you do pay a bit more for more reliable usage tier, for a task that requires it. It also shows you exactly how much it's spent at any point.

fcoury5• 1 year ago

... usage limits that would make it very difficult to use for Claude Code.

Genuinely interested: how's so?

zomglings13• 1 year ago

The writing of edits and code directly on my machine is something that happens on the client side. I don't see why that usage would be subject to anything but one-time billing or how it puts any strain on Anthropic's infrastructure.

ghuntley8• 1 year ago

$200/month isn't that much. Folks, I'm hanging around with are spending $100 USD to $500 USD daily as the new norm as a cost of doing business and remaining competitive. That might seem expensive, but it's cheap... https://ghuntley.com/redlining

m00dy2• 1 year ago

Seriously? That’s wild. What kind of CS field could even handle that kind of daily spend for a bunch of people?

sbszllr2• 1 year ago

All can be true depending on the business/person:

1. My company cannot justify this cost at all.

2. My company can justify this cost but I don't find it useful.

3. My company can justify this cost, and I find it useful.

4. I find it useful, and I can justify the cost for personal use.

5. I find it useful, and I cannot justify the cost for personal use.

That aside -- 200/day/dev for a "nice to have service that sometimes makes my work slightly faster" is much in the majority of the world.

oytis• 1 year ago

When should we expect to see the amazing products these super-competitive businesses are developing?

zomglings13• 1 year ago

$100/day seems reasonable as an upper-percentile spend per programmer. $500/day sounds insane.

A 2.5 hour session with Claude Code costs me somewhere between $15 and $20. Taking $20/2.5 hours as the estimate, $100 would buy me 12.5 hours of programming.

replwoacause2• 1 year ago

Their API billing in general is hostile to users. I switched completely to Gemini for this reason and haven’t looked back.

dist-epoch4• 1 year ago

Claude.ai/Desktop is priced based on average user usage. If you have 1 power user sending 1000 requests per day, and 99 sending 5, many even none, you can afford having a single $10/month plan for everyone to keep things simple.

But every Claude Code user is a 1000 requests per day user, so the economics don't work anymore.

fcoury5• 1 year ago

Well, take that into consideration then. Just make it an option. Instead of getting 1000 requests per day with code, you get 100 on the $10/month plan, and then let users decide whether they want to migrate to a higher tier or continue using the API model.

I am not saying Claude should stop making money, I'm just advocating for giving users the value of getting some Code coverage when you migrate from the basic plan to the pro or max.

Does that make sense?

zomglings13• 1 year ago

I would accept a higher-priced plan (which covered both my use of Claude.ai/Claude Desktop AND my use of Claude Code).

Anthropic make it seem like Claude Code is a product categorized like Claude Desktop (usage of which gets billed against your Claude.ai plan). This is how it signs off all its commits:

     Generated with [Claude Code](https://claude.ai/code)

At the very least, this is misleading. It misled me.

Once I had purchased the $200/month plan, I did some reading and quickly realized that I had been too quick to jump to conclusions. It still left me feeling like they had pulled a fast on one me.

twalkz• 1 year ago

I’ve been using codemcp (https://github.com/ezyang/codemcp) to get “most” of the functionality of Claude code (I believe it uses prompts extracted from Claude Code), but using my existing pro plan.

It’s less autonomous, since it’s based on the Claude chat interface, and you need to write “continue” every so often, but it’s nice to save the $$

zomglings13• 1 year ago

Thanks, makes sense that an MCP server that edits files is a workaround to the problem.

fcoury5• 1 year ago

Just tried it and it's indeed very good, thanks for mentioning it! :-)

jdance• 1 year ago

This would put anthropic in the business of minimizing the context to increase profits, same as Cursor and others who cheap out on context and try to RAG etc. Which would quickly make it worse, so I hope they stay on api pricing

Some base usage included in the plan might be a good balance

zomglings13• 1 year ago

You know, I wouldn't mind if they just applied the API pricing after Claude Code ran through the plan limits.

It would definitely get me to use it more.

visarga• 1 year ago

Yeah, tried it for a couple of minutes, $0.31, quickly stopped and moved away.

m00dy2• 1 year ago

well, the best practice is to use gemini 2.5 pro instead :)

replwoacause2• 1 year ago

Yep I learned this the hard way after racking up big bills just using Sonnet 3.7 in my IDE. Gemini is just as good (and not nearly as willing to agree with every dumb thing I say) and it’s way cheaper.

xpe8• 1 year ago

Gemini is ... way cheaper.

Yep. Here are the API pricing numbers for Gemini vs Claude. All per 1M tokens.

1. Gemini 2.5: in: $0.15; out: $0.60 non-thinking or $3.50 thinking

2. Claude 3.7: in: $3.00; out: $15

[1] https://ai.google.dev/gemini-api/docs/pricing [2] https://www.anthropic.com/pricing#api

sbszllr2• 1 year ago

The issue with many of these tips is that they require you use to claude code (or codex cli, doesn't matter) to spend way more time in it, feed it more info, generate more outputs --> pay more money to the LLM provider.

I find LLM-based tools helpful, and use them quite regularly but not 20 bucks+, let alone 100+ per month that claude code would require to be used effectively.

dist-epoch4• 1 year ago

let alone 100+ per month that claude code would require

I find this argument very bizarre. $100 is pay for 1-2 hours of developer time. Doesn't it save at least that much time in a whole month?

owebmaster4• 1 year ago

No, it doesn't. If you are still looking for product market fit, it is just cost.

After 2 years of GPT4 release, we can safely say that LLMs don't make finding PMF that much easier nor improve general quality/UX of products, as we still see a general enshittification trend.

If this spending was really game-changing, ChatGPT frontend/apps wouldn't be so bad after so long.

nrvn• 1 year ago

what happened to the "$5 is just a cup o' coffee" argument? Are we heading towards the everything-for-$100 land?

On a serious note, there is no clear evidence that any of the LLM-based code assistants will contribute to saving developer time. Depends on the phase of the project you are in and on a multitude of factors.

panny3• 1 year ago

Just a few days ago Cursor saved a lot of developer time by encouraging all the customers to quit using a product.

https://news.ycombinator.com/item?id=43683012

Developer time "saved" indeed ;-)

rpastuszak• 1 year ago

So sad that people are happy to spend 100$ pd on a tool like this, and we're so unlikely (in general) to pay $5 to an author of an article/blog posts that possibly saved you the same amount of time.

(I'm not judging a specific person here, this is more of a broad commentary regarding our relationship/sense of responsibility/entitlement/lack of empathy when it comes to supporting other people's work when it helps us)

ramoz• 1 year ago

Interesting, I have $100 days with Claude Code. Beyond effective.

bugglebeetle2• 1 year ago

Claude Code works fairly well, but Anthropic has lost the plot on the state of market competition. OpenAI tried to buy Cursor and now Windsurf because they know they need to win market share, Gemini 2.5 pro is better at coding than their Sonnet models, has huge context and runs on their TPU stack, but somehow Anthropic is expecting people to pay $200 in API costs per functional PR costs to vibe code. Ok.

owebmaster4• 1 year ago

but somehow Anthropic is expecting people to pay $200 in API costs per functional PR costs to vibe code. Ok.

Reading the thread, somehow people are paying. It is mindblowing how in place of getting cheaper, development just got more expensive for businesses.

tylersmith• 1 year ago

$200 per PR is significantly cheaper development than businesses are paying.

frainfreeze• 1 year ago

3.5 was amazing for code, and topped benchmarks for months. It'll take a while for other models to take over that mental space.

zoogeny4• 1 year ago

So I have been using Cursor a lot more in a vibe code way lately and I have been coming across what a lot of people report: sometimes the model will rewrite perfectly working code that I didn't ask it to touch and break it.

In most cases, it is because I am asking the model to do too much at once. Which is fine, I am learning the right level of abstraction/instruction where the model is effective consistently.

But when I read these best practices, I can't help but think of the cost. The multiple CLAUDE.md files, the files of context, the urls to documentation, the planning steps, the tests. And then the iteration on the code until it passes the test, then fixing up linter errors, then running an adversarial model as a code review, then generating the PR.

It makes me want to find a way to work at Anthropic so I can learn to do all of that without spending $100 per PR. Each of the steps in that last paragraph is an expensive API call for us ISV and each requires experimentation to get the right level of abstraction/instruction.

I want to advocate to Anthropic for a scholarship program for devs (I'd volunteer, lol) where they give credits to Claude in exchange for public usage. This would be structured similar to creator programs for image/audio/video gen-ai companies (e.g. runway, kling, midjourney) where they bring on heavy users that also post to social media (e.g. X, TikTok, Twitch) and they get heavily discounted (or even free) usage in exchange for promoting the product.

istjohn• 1 year ago

Why do you think it's supposed to be cheap? Developers are expensive. Claude doesn't have to be cheap to make software development quicker and cheaper. It just has to be cheaper than you.

There are ways to use LLMs cheaply, but it will always be expensive to get the most out of them. In fact, the top end will only get more and more costly as the lengths of tasks AIs can successfully complete grows.

zoogeny4• 1 year ago

I am not implying in any sense a value judgement on cost. I'm stating my emotions at the realization of the cost and how that affects my ability to use the available tools in my own education.

It would be no different than me saying "it sucks university is so expensive, I wish I could afford to go to an expensive college but I don't have a scholarship" and someone then answers: why should it be cheap.

So, allow me the space to express my feelings and propose alternatives, of which scholarships are one example and creative programs are another. Another one I didn't mention would be the same route as universities force now: I could take out a loan. And I could consider it an investment loan with the idea it will pay back either in employment prospects or through the development of an application that earns me money. Other alternatives would be finding employment at a company willing to invest that $100/day through me, the limit of that alternative being working at an actual foundational model company for presumably unlimited usage.

And of course, I could focus my personal education on squeezing the most value for the least cost. But I believe the balance point between slightly useful and completely transformative usages levels is probably at a higher cost level than I can reasonably afford as an independent.

qudat• 1 year ago

It just has to be cheaper than you.

Not when you need an SWE in order for it to work successfully.

solatic• 1 year ago

It just has to be cheaper than you

There's an ocean of B2B SaaS services that would save customers money compared to building poor imitations in-house. Despite the Joel Test (almost 25 years old! craxy...) asking whether you buy your developers the best tools that money can buy, because they're almost invariably cheaper than developer salaries, the fact remains that most companies treat salaries as a fixed cost and everything else threatens the limited budget they have.

Anybody who has ever tried to sell developer tooling knows, you're competing with free/open-source solutions, and it aint a fair fight.

k__• 1 year ago

That's why I like Aider.

You can protect your files in a non-AI way: by simply not giving write access to Aider.

Also, apparently Aider is a bit more economic with tokens than other tools.

zoogeny4• 1 year ago

I haven't used Aider yet, but I see it show up on HN frequently recently (the last couple of days specifically).

I am hesitant because I am paying for Cursor now and I get a lot of model usage included within that monthly cost. I'm cheap, perhaps to a fault even when I could afford it, and I hate the idea of spending twice when spending once is usually enough. So while Aider is potentially cheaper than Claude Code, it is still more than what I am already paying.

I would appreciate any comments on people who have made the switch from Cursor to Aider. Are you paying more/less? If you are paying more, do you feel the added value is worth the additional cost? If you are paying less, do you feel you are getting less, the same or even more?

Wowfunhappy6• 1 year ago

So I have been using Cursor a lot more in a vibe code way lately and I have been coming across what a lot of people report: sometimes the model will rewrite perfectly working code that I didn't ask it to touch and break it.

I don't find this particularly problematic because I can quickly see the unnecessary changes in git and revert them.

Like, I guess it would be nice if I didn't have to do that, but compared to the value I'm getting it's not a big deal.

zoogeny4• 1 year ago

I agree with this in the general sense but of course I would like to minimize the thrash.

I have become obsessive about doing git commits in the way I used to obsess over Ctrl-S before the days of source control. As soon as I get to a point I am happy, I get the LLM to do a check-point check in so I can minimize the cost of doing a full directory revert.

But from a time and cost perspective, I could be doing much better. I've internalized the idea that when the LLM goes off the rails it was my fault. I should have prompted it better. So I am now consider: how do I get better faster? And the answer is I do it as much as I can to learn.

I don't just want to whine about the process. I want to use that frustration to help me improve, while avoiding going bankrupt.

vessenes2• 1 year ago

i think this is particularly claude 3.7 behavior - at least in my experience, it's ... eager. overeager. smarter than 3."6" but still, it has little chill. gemini is better; o3 better yet. I'm mostly off claude as a daily driver coding assistant, but it had a really long run - longest so far.

jasonjmcghee5• 1 year ago

Surprised that "controlling cost" isn't a section in this post. Here's my attempt.

---

If you get a hang of controlling costs, it's much cheaper. If you're exhausting the context window, I would not be surprised if you're seeing high cost.

Be aware of the "cache".

Tell it to read specific files (and only those!), if you don't, it'll read unnecessary files, or repeatedly read sections of files or even search through files.

Avoid letting it search - even halt it. Find / rg can have a thousands of tokens of output depending on the search.

Never edit files manually during a session (that'll bust cache). THIS INCLUDES LINT.

The cache also goes away after 5-15 minutes or so (not sure) - so avoid leaving sessions open and coming back later.

Never use /compact (that'll bust cache, if you need to, you're going back and forth too much or using too many files at once).

Don't let files get too big (it's good hygiene too) to keep the context window sizes smaller.

Have a clear goal in mind and keep sessions to as few messages as possible.

Write / generate markdown files with needed documentation using claude.ai, and save those as files in the repo and tell it to read that file as part of a question. I'm at about ~$0.5-0.75 for most "tasks" I give it. I'm not a super heavy user, but it definitely helps me (it's like having a super focused smart intern that makes dumb mistakes).

If i need to feed it a ton of docs etc. for some task, it'll be more in the few $, rather than < $1. But I really only do this to try some prototype with a library claude doesn't know about (or is outdated). For hobby stuff, it adds up - totally.

For a company, massively worth it. Insanely cheap productivity boost (if developers are responsible / don't get lazy / don't misuse it).

bugglebeetle2• 1 year ago

If I have to spend this much time thinking about any of this, congratulations, you’ve designed a product with a terrible UI.

jasonjmcghee5• 1 year ago

Some tools take more effort to hold properly than others. I'm not saying there's not a lot of room for improvement - or that the ux couldn't hold the users hand more to force things like this in some "assisted mode" but at the end of the day, it's a thin, useful wrapper around an llm, and llms require effort to use effectively.

I definitely get value out of it- more than any other tool like it that I've tried.

sqs3• 1 year ago

It's fundamentally hard. If you have an easy solution, you can go make a easy few billion dollars.

tetha• 1 year ago

Mh. Like, I'm deeply impressed what these AI assistants can do by now. But, the list in the parent comment there is very similar to my mental check-list of pair-programming / pair-admin'ing with less experienced people.

I guess "context length" in AIs is what I intuitively tracked with people already. It can be a struggle to connect the Zabbix alert, the ticket and the situation on the system already, even if you don't track down all the zabbix code and scripts. And then we throw in Ansible configuring the thing, and then the business requriements by more, or less controlled dev-teams. And then you realize dev is controlled by impossible sales-terms.

These are scope -- or I guess context -- expansions that cause people to struggle.

oxidant3• 1 year ago

Think about what you would do in an unfamiliar project with no context and the ticket

"please fix the authorization bug in /api/users/:id".

You'd start by grepping the code base and trying to understand it.

Compare that to, "fix the permission in src/controllers/users.ts in the function `getById`. We need to check the user in the JWT is the same user that is being requested"

djtango• 1 year ago

I have been quite skeptical of using AI tools and my experiences using them have been frustrating for developing software but power tools usually come with a learning curve while "good product" with clean simplified interface often results in reduced capability.

VIM, Emacs and Excel are obvious power tools which may require you to think but often produce unrivalled productivity for power users

So I don't think the verdict that the product has a bad UI is fair. Natural language interfaces is such a step up from old school APIs with countless flags and parameters

BeetleB3• 1 year ago

Oh wow. Reading your comment guarantees I'll never use Claude Code.

I use Aider. It's awesome. You explicitly specify the files. You don't have to do work to limit context.

boredtofears2• 1 year ago

Yeah, I tried CC out and quickly noticed it was spending $5+ for simple LLM capable tasks. I rarely break $1-2 a session using aider. Aider feels like more of a precision tool. I like having the ability to manually specify.

I do find Claude Code to be really good at exploration though - like checking out a repository I'm unfamiliar with and then asking questions about it.

Jerry2• 1 year ago

I use Aider. It's awesome.

What do you use for the model? Claude? Gemini? o3?

LeafItAlone• 1 year ago

Aider is a great tool. I do love it. But I find I have to do more with it to get the same output as Claude Code (no matter what LLM I used with Aider). Sure it may end up being cheaper per run, but not when my time is factored in. The flip side is I find Aider much easier to limit.

simonw2• 1 year ago

With Claude Code you can at least type "/code" at any point to see how much it's spent, and it will show you when you end a session (with Ctrl+C) too.

The output of /cost looks like this:

  > /cost 
    ⎿  Total cost: $0.1331
       Total duration (API): 1m 13.1s
       Total duration (wall): 1m 21.3s

jjallen• 1 year ago

Not having to specify files is a humongous feature for me. Having to remember which file code is in is half the work once you pass a certain codebase size.

aitchnyu2• 1 year ago

After switching to Aider, I realized the other tools have been playing elaborate games to choose cheaper models and to limit files and messages in context, both of which increase their bills.

datavirtue3• 1 year ago

GitHub copilot follows your context perfectly. I don't have to tell it anything about files. I tried this initially and it just screwed up the results.

xpe8• 1 year ago

GitHub copilot follows your context perfectly. I don't have to tell it anything about files. I tried this initially and it just screwed up the results.

Just to make sure we're on the same page. There are two things in play. First, a language model's ability to know what file you are referring to. Second, an assistant's ability to make sure the right file is in the context window. In your experience, how does Claude Code compare to Copilot w.r.t (1) and (2)?

pclmulqdq• 1 year ago

It's interesting that this is a problem for people because I have never spent more than about $0.50 on a task with Claude Code. I have pretty good code hygiene and I tell Claude what to do with clear instructions and guidelines, and Claude does it. I will usually go through a few revisions and then just change anything myself if I find it not quite working. It's exactly like having an eager intern.

kiratp• 1 year ago

The productivity boost can be so massive that this amount of fiddling to control costs is counterproductive.

Developers tend to seriously underestimate the opportunity cost of their own time.

Hint - it’s many multiples of your total compensation broken down to 40 hour work weeks.

pizza• 1 year ago

Hard agree. Whether it's 50 cents or 10 dollars per session, I'm using it to get work done for the sake of quickly completing work that aims to unblock many orders of magnitude more value. But in so far as cheaper correct sessions correlate with sessions where the problem solving was more efficient anyhow, they're fairly solid tips.

Aurornis2• 1 year ago

The cost of the task scales with how long it takes, plus or minus.

Substitute “cost” with “time” in the above post and all of the same tips are still valuable.

I don’t do much agentic LLM coding but the speed (or lack thereof) was one of my least favorite parts. Using any tricks that narrow scope, prevent reprocessing files over and over again, or searching through the codebase are all helpful even if you don’t care about the dollar amount.

jillesvangurp• 1 year ago

Exactly. I've been using the chat gpt desktop app not because of the model quality but because of the UX. It basically seamlessly integrates with my IDEs (intellij and vs code). Mostly I just do stuff like select a few lines, hit option+shift+1, and say something like "fix this". Nice short prompt and I get the answer relatively quickly. Option+shift+1 opens chat gpt with the open file already added to the context. It sees what lines are selected. And it also sees the output of any test runs on the consoles. So just me saying "fix this" now has a rich context that I don't need to micromanage.

Mostly I just use the 4o model instead of the newer better models because it is faster. It's good enough mostly and I prefer getting a good enough answer quickly than the perfect answer after a few minutes. Mostly what I ask is not rocket science so perfect is the enemy of good here. I rarely have to escalate to better models. The reasoning models are annoyingly slow. Especially when they go down the wrong track, which happens a lot.

And my cost is a predictable 20$/month. The downside is that the scope of what I can ask is more limited. I'd like it to be able to "see" my whole code base instead of just 1 file and for me to not have to micro manage what the model looks at. Claude can do that if you don't care about money. But if you do, you are basically micro managing context. That sounds like monkey work that somebody should automate. And it shouldn't require an Einstein sized artificial brain to do that.

There must be people that are experimenting with using locally running more limited AI models to do all the micromanaging that then escalate to remote models as needed. That's more or less what Apple pitched for Apple AI at some point. Sounds like a good path forward. I'd be curious to learn about coding tools that do something like that.

In terms of cost, I don't actually think it's unreasonable to spend a few hundred dollars per month on this stuff. But I question the added value over the 20$ I'm spending. I don't think the improvement is 20x better. more like 1.5x. And I don't like the unpredictability of this and having to think about how expensive a question is going to be.

I think a lot of the short term improvement is going to be a mix of UX and predictable cost. Currently the tools are still very clunky and a bit dumb. The competition is going to be about predictable speed, cost and quality. There's a lot of room for improvement here.

charlie02• 1 year ago

If this is true, why isn't our compensation scaling with the increases in productivity?

jjmarr2• 1 year ago

I don't think about controlling cost because I price my time at US$40/h and virtually all models are cheaper than that (with the exception of o1 or Gemini 2.5 pro).

If I spend $2 instead of $0.50 on a session but I had to spend 6 minutes thinking about context, I haven't gained any money.

owebmaster4• 1 year ago

Important to remind people this is only true if you have a profitable product, otherwise you’re spending money you haven’t earned.

jasonjmcghee5• 1 year ago

If you do it a bit, it just becomes habit / no extra time or cognitive load.

Correlation or causation aside, the same people I see complain about cost, complain about quality.

It might indicate more tightly controlled sessions may also produce better results.

Or maybe it's just people that tend to complain about one thing, complain about another.

chewz• 1 year ago

My attempt is - Do not use Claude Code at all, it is terrible tool. It is bad at almost everything starting with making simple edits to files.

And most of all Claude Code is overeager to start messing with your code and run unnecessary $$ instead of making sensible plan.

This isn't problem with Claude Sonnet - it is fundamnetal problem with Claude Code.

winrid4• 1 year ago

I pretty much one shot a scraper from an old Joomla site with 200+ articles to a new WP site, including all users and assets, and converting all the PDFs to articles. It cost me like $3 in tokens.

gundmc• 1 year ago

Never edit files manually during a session (that'll bust cache). THIS INCLUDES LINT

Yesterday I gave up and disabled my format-on-save config within VSCode. It was burning way too many tokens with unnecessary file reads after failed diffs. The LLMs still have a decent number of failed diffs, but it helps a lot.

irthomasthomas• 1 year ago

I assume they use a conversation, so if you compress the prompt immediately you should only break cache once, and still hit cache on subsequent prompts?

So instead of Write Hit Hit Hit

It's Write Write Hit Hit Hit

sagarpatil2• 1 year ago

If I have to be so cautious while using a tool might as well write the code myself lol. I’ve used Claude Code extensively and it is one of the best AI IDE. It just gets things done. The only downside is the cost. I was averaging $35-$40/day. At this cost, I’d rather just use Cursor/Windsurf.

0x696C69613• 1 year ago

I mostly work in neovim, but I'll open cursor to write boilerplate code. I'd love to use something cli based like Claude Code or Codex, but neither of them implement semantic indexing (vector embeddings) the way Cursor does. It should be possible to implement an MCP server which does this, but I haven't found a good one.

isaksamsten• 1 year ago

I use a small plugin I’ve written my self to interact with Claude, Gemini 2.5 pro or GPT. I’ve not really seen the need for semantic searching yet. Instead I’ve given the LLM access to LSP symbol search, grep and the ability to add files to the conversation. It’s been working well for my use cases but I’ve never tried Cursor so I can’t comment on how it compares. I’m sure it’s not as smooth though. I’ve tried some of the more common Neovim plugins and for me it works better, but the preference here is very personal. If you want to try it out it’s here: https://github.com/isaksamsten/sia.nvim

sqs3• 1 year ago

Tool-calling agents with search tools do very well at information retrieval tasks in codebases. They are slower and more expensive than good RAG (if you amortize the RAG index over many operations), but they're incredibly versatile and excel in many cases where RAG would fall down. Why do you think you need semantic indexing?

0x696C69613• 1 year ago

Why do you think you need semantic indexing?

Unfortunately I can only give an anecdotal answer here, but I get better results from Cursor than the alternatives. The semantic index is the main difference, so I assume that's what's giving it the edge.

xpe8• 1 year ago

Good point. I largely work in Zed -- looks like it had semantic search for a while but is working on a redesign https://github.com/zed-industries/zed/issues/9564

Wowfunhappy6• 1 year ago

Use /clear to keep context focused

The only problem is that this loss is permanent! As far as I can tell, there's no way to go back to the old conversation after a `/clear`.

I had one session last week where Claude Code seemed to have become amazingly capable and was implementing entire new features and fixing bugs in one-shot, and then I ran `/clear` (by accident no less) and it suddenly became very dumb.

jasonjmcghee5• 1 year ago

They've worked to improve this with "memories" (hash symbol to "permanently" record something - you can edit later if you want).

And there's CLAUDE.md. it's like cursorrules. You can also have it modify it's own CLAUDE.md.

zomglings13• 1 year ago

You can ask it to store its current context to a file, review the file, ask it to emphasize or de-emphasize things based on your review, and then use `/clear`.

Then, you can edit the file at your leisure if you want to.

And when you want to load that context back in, ask it to read the file.

Works better than `/compact`, and is a lot cheaper.

Wowfunhappy6• 1 year ago

Neat, thanks, I had no idea!

Edit: It so happens I had a Claude Code session open in my Terminal, so I asked it:

    Save your current context to a file.

Claude produced a 91 line md file... surely that's not the whole of its context? This was a reasonably lengthy conversation in which the AI implemented a new feature.

datavirtue3• 1 year ago

Compared to my experience with the free GitHub Copilot in VS Code it sounds like you guys are in a horse and buggy.

andrewstuart• 1 year ago

I’m too scared of the cost to use this.

xpe8• 1 year ago

You can set spend limits https://docs.anthropic.com/en/api/rate-limits

LADev• 1 year ago

This is so helpful!

fallinditch• 1 year ago

I'm wondering how much of the techniques described in this blog post can be used in an IDE like Windsurf or Cursor with Claude Sonnet?

My 2 cents on value for money and effectiveness of Claude vs Gemini for coding:

I've been using Windsurf, VS Code and the new Firebase Studio. The Windsurf subscription allowance for $15 per month seems adequate for reasonable every day use. I find Claude Sonnet 3.7 performs better for me than Gemini 2.5 pro experimental.

I still like VS Code and its way of doing things, you can do a lot with the standard free plan.

With Firebase Studio, my take is that it should good for building and deploying simple things that don't require much developer handholding.

flashgordon4• 1 year ago

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 interesting aspect of doing it this way is being able to break up problems and concerns. For example in this case I only worked on fe without any backend and flushed it out before starting on an backend.

xpe8• 1 year ago

by fe the poster means FE (front-end)

flashgordon4• 1 year ago

Sorry yes. I should have clarified that.

neodypsis3• 1 year ago

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

vessenes2• 1 year ago

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.

flashgordon4• 1 year ago

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

simonw2• 1 year ago

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" < "think hard" < "think harder" < "ultrathink." Each level allocates progressively more thinking budget for Claude to use.

I had a poke around and it's not a feature of the Claude model, it's specific to Claude Code. There's a "megathink" option too - it uses code that looks like this:

  let B = W.message.content.toLowerCase();
  if (
    B.includes("think harder") ||
    B.includes("think intensely") ||
    B.includes("think longer") ||
    B.includes("think really hard") ||
    B.includes("think super hard") ||
    B.includes("think very hard") ||
    B.includes("ultrathink")
  )
    return (
      l1("tengu_thinking", { tokenCount: 31999, messageId: Z, provider: G }),
      31999
    );
  if (
    B.includes("think about it") ||
    B.includes("think a lot") ||
    B.includes("think deeply") ||
    B.includes("think hard") ||
    B.includes("think more") ||
    B.includes("megathink")
  )
    return (
      l1("tengu_thinking", { tokenCount: 1e4, messageId: Z, provider: G }), 1e4
    );

Notes on how I found that here: https://simonwillison.net/2025/Apr/19/claude-code-best-pract...

orojackson2• 1 year ago

Not gonna lie: the "ultrathink" keyword that Sonnet 3.7 with thinking tokens watches for gives me "doubleplusgood" vibes in a hilarious but horrifying way.

4b11b4• 1 year ago

At this point should we get our first knob/slider on a language model... THINK

..as if we're operating this machine as analog synth

coffeebeqn• 1 year ago

A little bit of the old ultrathink with the boys

pyfon• 1 year ago

Weird code to have in a modern AI system!

Also 14 string scans seems a little inefficient!

Aurornis2• 1 year ago

14 checks through a string is entirely negligible relative to the amount of compute happening. Like a drop of water in the ocean.

anotherpaulg• 1 year ago

In aider, instead of “ultrathink” you would say:

  /thinking-tokens 32k

Or, shorthand:

  /think 32k

westoncb• 1 year ago

That's awesome, and almost certainly an Unreal Tournament reference (when you chain enough kills in short time it moves through a progression that includes "megakill" and "ultrakill").

orojackson2• 1 year ago

If they did, they left out the best one: "m-m-m-m-monsterkill"

Surely Anthropic could do a better job implementing dynamic thinking token budgets.

Quarrel2• 1 year ago

Ultrakill is from Quake :)

mr-karan2• 1 year ago

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.

gdudeman• 1 year ago

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.

nulld3v• 1 year ago

Waiting until I can tell it to use "galaxy brain".

NiloCK• 1 year ago

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.

carlmr• 1 year ago

LLMs with broad contextual capabilities shouldn't need to be guided in this manor.

I mean finding your way around a manor can be hard, it's easier in an apartment.

adastra22• 1 year ago

“think hard with a vengeance”

panny3• 1 year ago

Use Claude to interact with git

Are they saying Claude needs to do the git interaction in order to work and/or will generate better code if it does?

sagarpatil2• 1 year ago

It doesn’t need to. Its optional.

panny3• 1 year ago

I don't see how this is a best practice then. It seems like they are saying "Spend money on something easy to do, but can be catastrophic if the AI screws it up."

jwr• 1 year ago

I use Claude Code. I read the discussion here, and given the criticism, proceeded to try some of the other solutions that people recommended.

After spending a couple of hours trying to get aider and plandex to run (and then with Google Gemini 2.5 pro), my conclusion is that these tools have a long way to go until they are usable. The breakage is all over the place. Sure, there is promise, but today I simply can't get them to work reasonably. And my time is expensive.

Claude Code just works. I run it (even in a slightly unsupported way, in a Docker container on my mac) and it works. It does stuff.

PS: what is it with all "modern" tools asking you to "curl somewhere.com/somescript.sh | bash". Seriously? Ship it in a docker container if you can't manage your dependencies.

beefnugs• 1 year ago

Isn't this bad that every model company is making their own version of the IDE level tool?

Wasn't it clearly bad when facebook would get real close to buying another company... then decide naw, we got developers out the ass lets just steal the idea and put them out of business

bob1029• 1 year ago

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 difference being that the contractor can be held accountable in our human legal system and can be made to follow their "prompt" very strictly. The LLM has its own advantages, but they seem to be a subset since the human contractor can also utilize an LLM.

Those who get a lot of uplift out of the models are almost certainly using them in a cybernetic manner wherein the model is an integral part of an expert's thinking loop regarding the program/problem. Defining a pile of policies and having the LLM apply them to a codebase automatically is a significantly less impactful use of the technology than having a skilled human developer leverage it for immediate questions and code snippets as part of their normal iterative development flow.

If you've got so much code that you need to automate eyeballs over it, you are probably in a death spiral already. The LLM doesn't care about the terrain warnings. It can't "pull up".

stepbeek• 1 year ago

This matches well with my experience so far. It’s why the chat interface has remained my preference over autocomplete in an IDE.

ixaxaar• 1 year ago

These are effectively identical to outsourcing your product to someone like Infosys.

But in my experience, the user has to be better than an Infosys employee to know how to convey the task to the LLM and then verify iteratively.

So more like an experienced engg outsourcing work to a service company engg.

anamexis• 1 year ago

That’s exactly what they were saying.

charlie02• 1 year ago

The benefit of doing it like this is that I also get to learn from the LLM. It will surprise me from time to time about things I didn't know and it gives me a chance to learn and get better as well.

Terretta2• 1 year ago

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.

highfrequency• 1 year ago

Intriguing perspective! Could you elaborate on this with another paragraph or two?

We humans undervisualize until we see concrete results.

imafish2• 1 year ago

Why do people use Claude Code over e.g. Cursor or Windsurf?

consumer451• 1 year ago

I have not yet used Claude Code personally, but I believe that Cursor and Windsurf both optimize token usage by limiting how much of your code each prompt analyzes.

With Claude Code, all bets are off there. You get a better understanding of your code in each prompt, and the bill can rack up, what, 50x faster?

If I got really stuck on a problem involving many lines of code, I could see myself spinning up Claude Code for that one issue, then quickly going back to Windsurf.

submeta• 1 year ago

I love Claude Code. It just gets the job done where Cursor (even with Claude Sonnet 3.7) will get lost in changing files without results.

Did anyone have equal results with the „unofficial“ fork „Anon Kode“? Or with Roo Code with Gemini Pro 2.5?

kkukshtel• 1 year ago

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.

appleaday1• 1 year ago

This is a pretty desperate post imho.

ccarse• 1 year ago

What are these "subagents" this doc refers to?

Claude Code: Best practices for agentic coding

Related Stories