Brb cat on fire

8K posts

Brb cat on fire banner
Brb cat on fire

Brb cat on fire

@brbcatonfire

Building Agent platform | R&D in AI | BigCorpo Refugee | Playing vid games, usually on Linux. | Mocking tech and society with love. 日本語上手

No where เข้าร่วม Nisan 2009
638 กำลังติดตาม243 ผู้ติดตาม
ทวีตที่ปักหมุด
Brb cat on fire
Brb cat on fire@brbcatonfire·
The Zen of software engineering: If it's not tested, it's broken. If it's in prod, it will be broken. If it's broken, it's on a sat. If it takes down a good chunk of the world infrastructure, you're probably related to microsoft somehow.
English
2
4
26
10.4K
YURIKO TIGER 🐯🎧
YURIKO TIGER 🐯🎧@yurikotigra·
I need cuddles. I need good news.. I need something to warm my heart rn. PLEASE POST WITH YOUR PUPPIES/Animals for cheer me up (;-;)
English
73
12
540
6.2K
Brb cat on fire
Brb cat on fire@brbcatonfire·
These look like good weekend projects to create a wiki for your life.
Andrej Karpathy@karpathy

Farzapedia, personal wikipedia of Farza, good example following my Wiki LLM tweet. I really like this approach to personalization in a number of ways, compared to "status quo" of an AI that allegedly gets better the more you use it or something: 1. Explicit. The memory artifact is explicit and navigable (the wiki), you can see exactly what the AI does and does not know and you can inspect and manage this artifact, even if you don't do the direct text writing (the LLM does). The knowledge of you is not implicit and unknown, it's explicit and viewable. 2. Yours. Your data is yours, on your local computer, it's not in some particular AI provider's system without the ability to extract it. You're in control of your information. 3. File over app. The memory here is a simple collection of files in universal formats (images, markdown). This means the data is interoperable: you can use a very large collection of tools/CLIs or whatever you want over this information because it's just files. The agents can apply the entire Unix toolkit over them. They can natively read and understand them. Any kind of data can be imported into files as input, and any kind of interface can be used to view them as the output. E.g. you can use Obsidian to view them or vibe code something of your own. Search "File over app" for an article on this philosophy. 4. BYOAI. You can use whatever AI you want to "plug into" this information - Claude, Codex, OpenCode, whatever. You can even think about taking an open source AI and finetuning it on your wiki - in principle, this AI could "know" you in its weights, not just attend over your data. So this approach to personalization puts *you* in full control. The data is yours. In Universal formats. Explicit and inspectable. Use whatever AI you want over it, keep the AI companies on their toes! :) Certainly this is not the simplest way to get an AI to know you - it does require you to manage file directories and so on, but agents also make it quite simple and they can help you a lot. I imagine a number of products might come out to make this all easier, but imo "agent proficiency" is a CORE SKILL of the 21st century. These are extremely powerful tools - they speak English and they do all the computer stuff for you. Try this opportunity to play with one.

English
0
0
0
14
Brb cat on fire
Brb cat on fire@brbcatonfire·
This is so much cope. Just cut your losses, shut the company down. If YC is dropping you for exposed fraud and IP theft, you just want people to forget who you are at this point. Not take more photos and reminance. 30 under 30 is undefeated in socialpath narcissists.
Selin Kocalar@kocalars

YC and Delve have parted ways. I still remember the day we took our YC interview at MIT. We’re so grateful to the community and every founder friend we’ve made. We'll continue to support every young founder striving to make the world a better place.

English
0
0
1
134
Brb cat on fire
Brb cat on fire@brbcatonfire·
@ChibiReviews When they make enough money as a manga artist they go back to doing what they love. Masamune Shirow is living his dream.
English
0
1
11
1.9K
Chibi Reviews
Chibi Reviews@ChibiReviews·
Fun Fact: Most Manga Authors have written hentai. Before they made their popular series. The fact I made people crash out by revealing the 100 Girlfriends author wrote Shotacon Hentai is fucking hilarious
Chibi Reviews tweet media
English
122
329
5.4K
116.4K
The Rabbit Hole
The Rabbit Hole@TheRabbitHole·
Hollywood went Woke. Now it’s going broke.
The Rabbit Hole tweet media
English
1.4K
2.6K
14K
14.6M
Brb cat on fire
Brb cat on fire@brbcatonfire·
アメ日本 Xが我々のサイバーパンクの未来の始まりだ
Brb cat on fire tweet media
日本語
0
0
1
40
Brb cat on fire
Brb cat on fire@brbcatonfire·
@LilPeepyX @Grummz LA at least has the Koreans lowering the price down for their sushi and the quality is decent. When i lived in London, the Chinese put out some pretty bad slop and called it "japanese fusion" too. Europe in general had bad japanese outside of the French Vietnamese sushi places
English
0
0
0
43
Lil Peepy
Lil Peepy@LilPeepyX·
@brbcatonfire @Grummz Way more than half. There are barely any Japanese restaurant owners in the Midwest but there are still tons of Chinese-owned “Japanese Fusion” restaurants. They invoke the name of Japan to make money but they serve cheap quality food. It’s moderately insulting.
English
1
0
3
61
Grummz
Grummz@Grummz·
X should permaban all location fakers pretending to be Japanese.
English
137
565
7.9K
88.3K
Brb cat on fire
Brb cat on fire@brbcatonfire·
@imyouhu Anthropic has such shitty programers relative to their researchers. And academics are terrible programmers too.
English
1
0
15
1.6K
CatFly
CatFly@imyouhu·
Claude Code 源码泄露事件后续越来越精彩了。 有人拿泄露的源码丢给 OpenAI 的 Codex 分析,竟然找到了 Claude Code 疯狂消耗 token 的元凶——autoCompact(自动上下文压缩)机制在失败后会无限重试,完全没有上限。据源码注释记录,曾有会话连续失败高达 3272 次。 修复方法简单到离谱:加一个 MAX_CONSECUTIVE_AUTOCOMPACT_FAILURES = 3 的限制,连续失败 3 次就停止重试。三行代码,搞定。 打完补丁后,这位老哥表示使用额度恢复正常了——之前被吐槽的"用两下就触发限速",很可能有一部分就是这个 bug 在背后偷偷烧 token。 仓库地址放下面了。
Lydia Hallie ✨@lydiahallie

We're aware people are hitting usage limits in Claude Code way faster than expected. Actively investigating, will share more when we have an update!

中文
129
731
6.3K
1M
frye
frye@___frye·
the magic of talking to japanese twitter is that auto-translation tools can’t quite capture irony or meme speak everyone has to strip away the layers of internet poisoning and talk sincerely again
English
28
181
3.8K
42.9K
Brb cat on fire
Brb cat on fire@brbcatonfire·
@paulg Thats because AI raised the floor when most people are retards. The best posters are still better than AI for now.
English
0
0
2
35
Paul Graham
Paul Graham@paulg·
Sadly, one way I now recognize fake AI-generated replies is that AIs write punchier sentences than most ordinary humans.
English
666
64
2.9K
207.2K