[flagged]
Can you please not gut your comments by deleting the content, once they have replies?
It destroys the context for readers, making the thread worse for everyone.
How I use Claude Code to implement new features in an existing complex codebase
61–70 of 71 posts
Re: How I use Claude Code to implement new features in an existing complex codebase
#62Earlier quoted context omitted.
.
[flagged]
This crosses into personal attack, which is not allowed here. Please don't post like this again.
In addition, please note these guidelines:
"When disagreeing, please reply to the argument instead of calling names. 'That is idiotic; 1 + 1 is 2, not 3' can be shortened to '1 + 1 is 2, not 3."
"Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."
"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."
https://news.ycombinator.com/newsguidelines.html
Edit: this is unfortunately a bit of a pattern with your account (e.g. https://news.ycombinator.com/item?id=44665780), and we had to warn you about this at least once before. If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
Re: How I use Claude Code to implement new features in an existing complex codebase
#63That many sites about AI look a lot like scam sites ("make money fast") is a real turn-off -> https://www.sabrina.dev Maybe it does have some valuable content but I will just close it.
I agree. I remember a year ago some a submission on this forum by Sabrina where she was evaluating LLMs on SAT math problems. The post was called like “ChatGPT-4o vs. Math”. I thought it was pretty insightful, but now that post is deleted from her blog and it seems like the blog as a whole is leaning into attracting people who are nontechnical and want to use it to generate social media content. Weird vibe in the comments of her posts and videos too.
Re: How I use Claude Code to implement new features in an existing complex codebase
#64Re: How I use Claude Code to implement new features in an existing complex codebase
#65I stumbled upon the author’s command `qnew` to read Claude.md. Why would you need that? > CLAUDE.md is a special file that Claude automatically pulls into context when starting a conversation. https://www.anthropic.com/engineering/claude-code-best-pract...
That confused me as well. The whole point of CLAUDE.md is that it is automatically injected into the rolling context of any Claude interaction/thread. That CLAUDE.md file that they've posted is also HUGE - Anthropic themselves recommend keeping it on the more concise side. If you need more, consider just creating dedicated subagents for "UI/UX reviewer", "search", etc.
My CLAUDE.md has references to other documentation
Like "for guidance on Go library selection refer to docs/go.md"
That way the main file stays compact.
Re: How I use Claude Code to implement new features in an existing complex codebase
#66Earlier quoted context omitted.
Can you please not gut your comments by deleting the content, once they have replies? It destroys the context for readers, making the thread worse for everyone.
That’s fair, I shared things that weren’t in a state to communicate. I shouldn’t have commented at all.
Appreciated! And yes, that happens sometimes.
Re: How I use Claude Code to implement new features in an existing complex codebase
#67That many sites about AI look a lot like scam sites ("make money fast") is a real turn-off -> https://www.sabrina.dev Maybe it does have some valuable content but I will just close it.
I agree. I remember a year ago some a submission on this forum by Sabrina where she was evaluating LLMs on SAT math problems. The post was called like “ChatGPT-4o vs. Math”. I thought it was pretty insightful, but now that post is deleted from her blog and it seems like the blog as a whole is leaning into attracting people who are nontechnical and want to use it to generate social media content. Weird vibe in the com…
I migrated from beehiiv to substack a few months ago but haven't migrated some of my older posts
Re: How I use Claude Code to implement new features in an existing complex codebase
#68Earlier quoted context omitted.
That confused me as well. The whole point of CLAUDE.md is that it is automatically injected into the rolling context of any Claude interaction/thread. That CLAUDE.md file that they've posted is also HUGE - Anthropic themselves recommend keeping it on the more concise side. If you need more, consider just creating dedicated subagents for "UI/UX reviewer", "search", etc.
My CLAUDE.md has references to other documentation Like "for guidance on Go library selection refer to docs/go.md" That way the main file stays compact.
subagents were not yet released at the time of my blog post
Re: How I use Claude Code to implement new features in an existing complex codebase
#69Earlier quoted context omitted.
The grifters have entered the arena.
I mean, the grifters have started it. They grifted for way too much money without ANY plan on how to actually get money back to the investors. Including the dubious benefits, just promising a "quick fix". That is a bit over the top, but the core statement kind of stands.
I'm the writer of the blog post. I don't sell ai courses, consulting, coaching, or anything like that. But I do use claude code to help me build my SaaS app as a bootstrapped solopreneur. I also sold an AI startup in early 2021, which I founded back in 2014 when the hyped term was "Big Data". I've been around a few hype cycles now..
Re: How I use Claude Code to implement new features in an existing complex codebase
#70Earlier quoted context omitted.
I mostly do the same. My workflow is usually like this: 1. Supply CC with a simple user story with light spec / business logic, asking it to write a specification for the feature and ask me any questions that need to be answered to make it complete. 2. Build up the context with all relevant files that I want to use. In Ruby on Rails it means refercing models, views and controllers that show the structure we are going…
Nice to see!! I wrote a longer post about this: https://open.substack.com/pub/boliv/p/claude-code-usage-patt...
Not sure the complexity of your app but mine's pretty complex with many daily active users. One-shot rarely works unless it's a simple feature/chore and I tell CC exactly what to do, which functions to touch, and how (which defeats the purpose).