stopping your agents from going rogue @prismor_dev | OSS (4k stars) | @fdotinc

Joined August 2009
148 Photos and videos
Wanted to do something different this weekend, so i built a little corner of the internet where you can turn ANY website into its own animated pixel-art mascot 🐕 go try it out, it's free: webassets.art
1
6
4,877
thanks @UseCorgi for giving me the space to build this today
33
Fable is out AGAINN and while it is the most capable model. We are trying to make sure your coding agents are much more secure we have prismor all opensource, would love to have all of you guys try this and give feedback: github.com/PrismorSec/prismo…
3
1
10
505
It sucks building in the security space and using Fable
4
10
277
One week in SF and I'm already surrounded by incredible people building incredible things One lesson I keep coming back to from Founders Inc. : start with the skateboard, not the car. Ride it until it breaks. Learn what users actually need. Then build the bike and then the car. If you're in SF, I'd love to connect 👋
1
19
462
Model routing is becoming the new infrastructure layer Is there any good solution for it??
1
3
126
Robo fight
2
11
1,925
Arnav Gupta reposted
Chrome 里居然藏了个能白嫖的 AI,很多人压根不知道。 最近的 Chrome 把 Google 的小模型 Gemini Nano 塞进了浏览器里。开个设置,不用联网、不用注册,直接在本地就能跑,模型还是装在你自己电脑上。 唯一的坑是:以前这玩意儿得开发者写代码才调得出来。 结果海外一个老哥(@_ar9av)做了个工具,把它包成跟 ChatGPT API 一样的调用方式,连 Ollama 这类本地软件都省了。 一条命令搞定: 🔗 npm start 跑完会起两样东西: 1️⃣ 浏览器里的聊天界面 2️⃣ 兼容 ChatGPT 的接口(只能本地用) 白嫖党狂喜。 github.com/Ar9av/gemini-nano…
3
8
19
5,402
Arnav Gupta reposted
x.com/i/article/206800530139…
3
1
4
3,135
dear X Algo, connect me with people joining Off Season 2 @fdotinc
3
12
9,235
x.com/_ar9av/status/20596307…
What made me applied to Off Season II @fdotinc ? -> @hthieblot lets go!
4
397
So apparently Google ships a Gemini Nano 4B LLM (context limit : 9216 tokens) baked into chrome I tried to expose it out as an OpenAI-compatible API for my local, so basically no API key, no external network calls, no need of ollama Demo and repo below ↓
23
78
963
151,046
Working with knowledge graphs again and the hybrid QA problem hasn't gone away embedding search finds fluffy neighbors, graph traversal finds precise neighbors you still need both and the fusion layer is where all the interesting failures happen
4
237
day 19 of reading one arxiv paper around AI every day and sharing what actually stuck Creative Machine (Concordia University) TLDR: generating creative-looking outputs and being genuinely creative are different things this paper draws the line and gives 10 requirements for crossing it, most current AI systems fail on the ones that matter most
1
2
170
the ethical implication is specific: a genuinely creative system doesn't just produce artifacts. it intervenes in situations and reshapes them. that means ethical constraints can't be a post-generation review who's affected, what harms are possible, what's out of bounds those have to shape what the system perceives, what it identifies as a conflict and which interventions it even considers
1
113