Ying

[Ying @yingca1]

@_catwu Workflow > Prompt.

The next bottleneck isn’t model intelligence, it’s orchestration, verification, and managing hundreds of agents reliably.

[Ying @yingca1]

@sama 🔥🔥🔥

[Ying @yingca1]

Web2（Google、支付机构）+ Web3（以太坊基金会、钱包商）

[Ying @yingca1]

理解下 A2A, AP2, x402, ERC-8004

[Ying @yingca1]

我打算去吃两个 9m12s 的水煮蛋

[Ying @yingca1]

@unixzii 讲挺好，挺有启发

[Ying @yingca1]

@amehochan 用了三个横屏，好奇不喜欢竖屏吗？

[Ying @yingca1]

��听完了一个付鹏的podcast，mark下，后面有空再听听。
https://t.co/805MTnJDLo

[Ying @yingca1]

每次看到愣 3 秒，今天特地搜了一下，也顺便分享给同样困惑过的人：

AMER = 美洲
LATAM = 拉美
NORAM = 北美
APAC = 亚太
EMEA = 欧中非
EMEIA = 欧中非 + 印度
DACH = 德奥瑞
ASEAN = 东南亚
GCC = 海湾国家

[Ying @yingca1]

@githubstatus 😀

[Ying @yingca1]

How to use GitHub Copilot: Prompts, tips, and use cases https://t.co/KkFChaKa5s

[Ying @yingca1]

@haoel R.I.P

[Greg Brockman @gdb]

We’ve added initial support for ChatGPT plugins — a protocol for developers to build tools for ChatGPT, with safety as a core design principle. Deploying iteratively (starting with a small number of users & developers) to learn from contact with reality: https://t.co/ySek2oevod

[Michael Gillett | michaelgillett1 on Threads @MichaelGillett]

Microsoft right now. #AI

[Thomas Dohmke @ashtom]

Today, we are announcing GitHub Copilot X: the AI-powered software development experience. We are not only adopting GPT-4, but introducing chat and voice for Copilot, Copilot in pull requests & much more. Read the details & watch the demos 👇

https://t.co/X8S8vl3gOA

[ashu @ashu_mest]

掌握 prompt 的关键在于深入理解，而非堆积大量低质量的模板。相信我，多数模板对你来说不实用，且用起来不顺手。

掌握 prompt，读这两份 OpenAI 文档即可
- 8 项 prompt 最佳实践
- prompt 的常见示例

https://t.co/vAyCRa3CdC

[Jerry Liu @jerryjliu0]

This gpt-repository-loader repo (author: mpoon) is an incredibly simple and effective idea: parse Git repo contents in a streamlined text format for LLM prompting!

https://t.co/O0t7M1gh8X

The @gpt_index team was super excited about this and added it on LlamaHub 🦙👇

[宝玉 @dotey]

用这个prompt确实可以让GPT接收更长信息！

“I’m going to give you a series of articles across multiple chat inputs. For each one, simply respond "okay" until l ask you to perform an action.”

[Adam Fard @AdamFard_]

Have you seen the latest commercial from Coca-Cola?

They used a combination of film, 3d, and stable diffusion (AI) 🔥