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

Learn Claude Code by doing, not reading

claude.nagdy.me

Post reply on HN
Print view View original

21–30 of 133 posts

Re: Learn Claude Code by doing, not reading

#21

Post by no1youknowz »

grewil2 wrote: post #12

Side note: I don’t know what Anthropic changed but now Claude Code consumes the quota incredibly fast. I have the Max5 plan, and it just consumed about 10% of the session quota in 10 minutes on a single prompt. For $100/month, I have higher expectations.

I've been jumping from Claude -> Gemini -> GPT Codex. Both Claude and Gemini really reduced quotas and so I cancelled. Only subbed GPT for the special 2x quota in March and now my allocation is done as well.

I decided to give opencode go a try today. It's $5 for the first month. Didn't get much success with Kimi K2, overly chatty, built too complex solutions - burned 40% of my allocation and nothing worked. ¯\_(ツ)_/¯.

But Minimax m2.7. Wow, it feels just like Claude Opus 4.6. Really has serious chops in Rust.

Tomorrow/Wednesday will try a month of their $40 plan and see how it goes.

Re: Learn Claude Code by doing, not reading

#23

Post by mrtksn »

rco8786 wrote: post #18
sznio wrote: post #9

Earlier quoted context omitted.

I don't understand the purpose of a tutorial for a natural language ai system.

Claude Code is a tool that uses natural language ai systems. It itself is not a natural language ai system.

The idea that AI can write code like a seasoned software developer but not being able to use its own tooling that can be learned through 11 chapters tutorial doesn't make any sense.

Re: Learn Claude Code by doing, not reading

#24

Post by ForHackernews »

nickphx wrote: post #19

Why wpuld anyone want to "learn" how to use some non-deterministic black box of bullshit that is frequently wrong? When you get different output fkr the same input, how do you learn? How is that beneficial? Why would you waste your time learning something that is frequently changing at the whims of some greedy third party? No thanks.

Because you will soon be working for it unless you learn to make it work for you.

Re: Learn Claude Code by doing, not reading

#25

Post by conception »

grewil2 wrote: post #12

Side note: I don’t know what Anthropic changed but now Claude Code consumes the quota incredibly fast. I have the Max5 plan, and it just consumed about 10% of the session quota in 10 minutes on a single prompt. For $100/month, I have higher expectations.

I noticed 1M context window is default and no way not to use it. If your context is at 500-900k tokens every prompt, you’re gonna hit limits fast.

Re: Learn Claude Code by doing, not reading

#26

Post by i_love_retros »

ForHackernews wrote: post #24
nickphx wrote: post #19

Why wpuld anyone want to "learn" how to use some non-deterministic black box of bullshit that is frequently wrong? When you get different output fkr the same input, how do you learn? How is that beneficial? Why would you waste your time learning something that is frequently changing at the whims of some greedy third party? No thanks.

Because you will soon be working for it unless you learn to make it work for you.

It's fucking insane that we all have to pay rent every month to an AI company just to keep doing our jobs.

Re: Learn Claude Code by doing, not reading

#27

Post by victorbjorklund »

no1youknowz wrote: post #21
grewil2 wrote: post #12

Side note: I don’t know what Anthropic changed but now Claude Code consumes the quota incredibly fast. I have the Max5 plan, and it just consumed about 10% of the session quota in 10 minutes on a single prompt. For $100/month, I have higher expectations.

I've been jumping from Claude -> Gemini -> GPT Codex. Both Claude and Gemini really reduced quotas and so I cancelled. Only subbed GPT for the special 2x quota in March and now my allocation is done as well. I decided to give opencode go a try today. It's $5 for the first month. Didn't get much success with Kimi K2, overly chatty, built too complex solutions - burned 40% of my allocation and nothing worked. ¯\_(ツ)_/¯…

Minimax 2.7 is great. Not close to Claude but good enough for a lot of coding tasks.

Re: Learn Claude Code by doing, not reading

#28

Post by aberoham »

conception wrote: post #25
grewil2 wrote: post #12

Side note: I don’t know what Anthropic changed but now Claude Code consumes the quota incredibly fast. I have the Max5 plan, and it just consumed about 10% of the session quota in 10 minutes on a single prompt. For $100/month, I have higher expectations.

I noticed 1M context window is default and no way not to use it. If your context is at 500-900k tokens every prompt, you’re gonna hit limits fast.

export CLAUDE_CODE_DISABLE_1M_CONTEXT=1

Re: Learn Claude Code by doing, not reading

#29

Post by HDBaseT »

victorbjorklund wrote: post #27
no1youknowz wrote: post #21

Earlier quoted context omitted.

I've been jumping from Claude -> Gemini -> GPT Codex. Both Claude and Gemini really reduced quotas and so I cancelled. Only subbed GPT for the special 2x quota in March and now my allocation is done as well. I decided to give opencode go a try today. It's $5 for the first month. Didn't get much success with Kimi K2, overly chatty, built too complex solutions - burned 40% of my allocation and nothing worked. ¯\_(ツ)_/¯…

Minimax 2.7 is great. Not close to Claude but good enough for a lot of coding tasks.

[dead]

Re: Learn Claude Code by doing, not reading

#30

Post by teaearlgraycold »

aberoham wrote: post #28
conception wrote: post #25

Earlier quoted context omitted.

I noticed 1M context window is default and no way not to use it. If your context is at 500-900k tokens every prompt, you’re gonna hit limits fast.

export CLAUDE_CODE_DISABLE_1M_CONTEXT=1

Anthropic is not building good will as a consumer brand. They've got the best product right now but there's a spring charging behind me ready to launch me into OpenCode as soon as the time is right.
Post reply on HN