Keep on to blur preview images; turn off to show them clearly
兄弟们，这个也太炸裂了🔥，别看什么 nano banana 了🍌 首个国人打造！效果直逼 Manus！专为 Agent 设计的 Memory 项目 - AContext 正式开源💥 - 存储 agent 对话、反馈、 artifacts - 自动提取 sop，优化 token 提高效果 - 让你的 agent 持续进化，比肩 manus https://t.co/coh9Url457
核心介绍如下
RT @indie_maker_fox: 🎉 MkDollar 开发进展 输入域名就可以抓取产品信息，然后你就可以管理产品的外链 你可以从外链库中轻松查找高DR+dofollow+无需登录的导航站，一个个去提交，并管理提交状态。 还在开发中，规划功能较多，敬请期待 🫡…
🔥 The best AI SaaS boilerplate - https://t.co/VyNtTs0jSX 🚀 The best directory boilerplate with AI - https://t.co/wEvJ1Dd8aR 🎉 https://t.co/bh1RxeERuY & https://t.co/zubXJCoY92 & https://t.co/tfQf8T7gGF
"AI-powered" has no appeal whatsoever. There are so many products that just shout out like "Hey, I'm using AI! I'm better than alternatives!" It's like "oh, I've got a Rasberrry Pi, so I made something with it". I'm always happy when a tool solves something, and it doesn't necessarily have to use AI.
Maker of a tech note-taking app https://t.co/xcprYEEMJQ. My indie dev journey 👉https://t.co/JbdJ58XbTN 🇯🇵@craftzdog Dad👨👩👧 UTC+9
I still maintain that reflective learning is the future of learning algorithms. This is related to but quite a bit richer than thinking about making value functions that work.
The problem with conventional supervised and reinforcement learning is that the system gets subjected to gradient updates it didn’t curate from experiments it didn’t design. If we’re all about “scaling thinking time”, shouldn’t the agent think much harder about how it learns?
RT @karakhanyanS: We have 3rd ad banner taken by @Bolt__AI by @daniel_nguyenx in https://t.co/RUe6DRnUMx
Just a funny dude building stuffs @Bolt__AI. Father, learner, maker. Expect memes.
RT @gallabytes: TPUs are not ASICs they're general purpose VLIW machines with wide af SIMD instructions & systolic array tensor cores.
🇦🇺 Co-founder: @AnswerDotAI & @FastDotAI ; Prev: professor @ UQ; Stanford fellow; @kaggle president; @fastmail/@enlitic/etc founder https://t.co/16UBFTX7mo