惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

月光博客
月光博客
D
Docker
腾讯CDC
J
Java Code Geeks
大猫的无限游戏
大猫的无限游戏
The Cloudflare Blog
Martin Fowler
Martin Fowler
MongoDB | Blog
MongoDB | Blog
博客园 - Franky
博客园 - 三生石上(FineUI控件)
Recent Announcements
Recent Announcements
F
Fortinet All Blogs
IT之家
IT之家
WordPress大学
WordPress大学
M
MIT News - Artificial intelligence
爱范儿
爱范儿
Microsoft Azure Blog
Microsoft Azure Blog
Vercel News
Vercel News
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
小众软件
小众软件
N
Netflix TechBlog - Medium
T
Tailwind CSS Blog
Engineering at Meta
Engineering at Meta
博客园 - 【当耐特】

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
Paul Graham (@paulg)
kklisura · 2026-06-18 · via Hacker News - Newest: "AI"

I met a founder today who said he writes 10,000 lines of code a day now thanks to AI. This is probably the limit case. He's a hotshot programmer, he knows AI tools very well, and he's talking about a 12 hour day. But he's not naive. This is not 10,000 lines of bug-filled crap.

Aug 7, 2025 · 2:59 AM UTC

1,313

508

12,096

2,852,705

He doesn't have any employees, and doesn't plan to hire any in the near future. Not because AI has made employees obsolete, but simply because he's so massively productive right now that he doesn't want to stop programming to spend time interviewing candidates.

131

82

3,440

293,543

This post is so out of touch I don’t even know what to say

16

5

1,771

135,845

So you've solved that problem by saying nothing.

24

2

1,117

108,970

paul. buddy. this is the opposite of a success story. you met a guy who tricked himself into thinking productivity means flooding a repo with 10,000 lines of code a day and your takeaway was "wow, what a wizard." no. what you saw was a man deep in the throes of AI assisted mania, mainlining synthetic velocity and calling it engineering. this is not craftsmanship. this is cope with autocomplete. and the worst part is you fell for it. what matters isn't how much code you generate. it's whether that code should exist at all. whether it's legible. whether it's real. lines of code is not a meaningful metric in software, never has been. AI can generate garbage that compiles at industrial scale. that doesn't mean we should applaud people who shovel it the fastest. you're watching a guy pour concrete onto a foundation made of jello and calling it a skyscraper because it’s tall and he built it fast. there is no way this person is producing 10k lines of code a day without bugs. this is slop paul and you know it.

37

18

1,013

77,253

I've known him for several years. I know he's a good programmer, and an expert in AI specifically and thus aware of the pitfalls. You really don't know what you're talking about here. I'm not trying to change your mind, just explaining why I'm blocking you.

43

2

246

47,751

10K LOC a day means there's no human reviewing that. So who's confirming the quality and more importantly, the security?

31

7

881

99,968

As it turns out I met another startup today that's using AI to do ongoing testing of changing codebases. (He's not using them yet though.)

36

6

309

91,323

can you imagine, not hiring an Editor to review your book?

5

1

96

60,433

Yes. In fact I can't imagine letting an editor mangle my writing.

9

6

304

55,734

Can you think of any reason why a founder might want to tell these tales to Paul Graham regardless of their truth?

2

90

23,743

In this case, no. YC already funded his company several years ago. We were just talking about the current state of AI-assisted programming.

3

28

21,042

I’m on line 702,244 today, keep up

11

9

630

26,637

I highly doubt it can be 10k lines of high-quality and bug free code. It also depends on what kind of code he wrote: e.g. frontend or backend, business logic or unit tests.

15

4

332

36,400

The best programmer I ever knew measured his productivity by how many lines of code he REMOVED from the code-base per day. So there's that.

10

9

491

16,304

your mom is 10,000 lines a day

1

3

163

16,020

Why measure code in lines?

8

103

12,551

Writing code was never the bottleneck - what the hell is someone even doing with 10k lines of code per day?

4

267

6,635

by the end of the year he’s at 3million LOC that’s age of empire online!! by the end of this decade 36 million !!!!! it’s the equivalent of windows XP codebase!!!!!!!

4

48

4,645

yeah, AI guys say a lot of things but lying keeps working for them so why would they stop

1

41

2,818

I work in cybersecurity and there are my favorite devs! They bring so many bugs that we would need to do multiple reviews just to make sure it has some normal level of security.

2

34

3,063

PG met Yacine?

1

31

9,454

As a general rule of thumb, I tend to believe the fewer lines of code to achieve a given result, the better the coder.

25

5,917

Lines of code written per day is a bad metric to measure productivity 😅 I can guarantee there will be bugs, if he says that’s not the case, I’m 99.99% sure he’s lying.

2

23

6,060

i cannot comprehend how you could have so many things to build

3

22

10,008

PER DAY? 😳 love the empowerment but code is such a liability. scaling chaos is an interesting strategy. less is usually more in the long run.

2

22

2,889

I can write 100,000 lines of code with AI that will compile in 8 hours. That being said I can write 800 lines with the help of AI that has 10x the value of those 100,000 lines.

1

21

2,428