跳转至

Readwise Daily｜2026-06-30

今日结论

今天的高价值内容集中在 AI、Programming。最值得优先处理的是《Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding》。整体建议是少追新闻，多沉淀能服务求职、英语/日语学习和 AI 工程实践的材料。今天只做一件事：读完 Top 1，并写下一个可执行行动。

快速概览

  • 输入

    45 条新增/更新内容

  • 值得读

    3 条 S 级，5 条可扫读

  • 主题分布

    AI: 31 · Programming: 6 · Business: 1 · Other: 7

  • 窗口

    2026-06-28T22:31:35.688Z → 2026-06-29T22:31:35.688Z

今日重点

1. 🤖 Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding

  • 主题：AI
  • 动作：READ
  • 分数：92/100
  • 作者：Simon Willison's Weblog
  • 链接：Reader / Source
  • 理由：和长期目标强相关，值得完整读

Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding This is an interesting new open weights (MIT licensed) model, the first model release from DeepReinforce.

[...] with variants including 9B Dense, 31B Dense, 35B MoE, and 397B MoE. Built on top of pretrained Gemma 4 and Qwen 3.5, it achieves state-of-the-art performance among open-source models of comparable size on coding benchmarks.

As far as I can tell the licenses of those underlying models is compatible with being used in this way - Gemma 4 is Apache 2.0 licensed (and not bound by the janky additional Gemma Terms of Use that afflicted the previous Gemma models) and Qwen 3.5 is Apache 2.0 licensed as well. I've been running the model using LM Studio and the ornith-1.0-35b-Q4_K_M.gguf (20GB) GGUF, hooked up to Pi. Initial impressions are very good - it seems to be able to run the agent harness over many tool calls in a proficient way. Here's a terminal session where I asked it to "find the code that decodes the actor cookie" and then "find the code that opens the insert dialog when thebutton is clicked" against a Datasette checkout, which it handled with ease. I also had it draw this pelican, which came out at 103 tokens/second:

It's a little bit mangled but the pelican is clearly a pelican. I couldn't find much information about DeepReinforce themselves. The earliest paper I could find from the was CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning from June 2025.

Tags: ai, generative-ai, local-llms, llms, qwen, pelican-riding-a-bicycle, gemma, llm-release, lm-studio

2. 🤖 US Grid Constraints: Towards 40GW+ of Behind-the-Meter Datacenter by 2028?

  • 主题：AI
  • 动作：READ
  • 分数：84/100
  • 作者：Hacker News
  • 链接：Reader / Source
  • 理由：和长期目标强相关，值得完整读

Today, the US grid is serving most datacenter load in the US, but we’re reaching a tipping point. As the insatiable demand for power of AI Labs and hyperscalers keeps accelerating, the grid simply can’t add capacity fast enough. That leaves Behind-The-Meter as the only way for th

3. 💻 JumpServer: Open-Source Privileged Access Management

  • 主题：Programming
  • 动作：READ
  • 分数：82/100
  • 作者：Hacker News
  • 链接：Reader / Source
  • 理由：和长期目标强相关，值得完整读

jumpserver/jumpserver An open-source PAM tool (Bastion Host) English · 简体中文 What is JumpServer? JumpServer is an open-source Privileged Access Management (PAM) tool that provides DevOps and IT teams with on-demand and secure access to SSH, RDP, Kubernetes, Database and RemoteApp

值得扫读

Back to blog How I learned to love the loop When Claude Code writes something that breaks, the reflex is to copy the error out of your terminal and paste it... - [Programming] A field guide to the modern front end for developers who hand-wrote HTML：有信息量，但不用深读

THE DESCENT · what happened to the frontend The last time you wrote a frontend, it looked like one of these —

  • <?= $it... - [AI] Why did one day of AI cost more than a month of servers?：有信息量，但不用深读

    Same old story: I’m running the SaaS our CFO shipped to production in two days. A non-engineer exec builds something fast with Claude Code, and the engineer... - [Business] European ISPs Want Rightsholders Held Accountable for Overblocking Damage：有信息量，但不用深读

    For years, rightsholders have pushed for broader site blocking orders, with no direct liability if these result in overblocking. EuroISPA, which represents o... - [AI] How AI Agents Manage Memory and Avoid Forgetfulness：有信息量，但不用深读

    When devs use AI to generate thousands of lines of unverified code, you risk a codebase slopocalypse.

    • 适合保存，暂不深读

    跳过

    • Count the number of Safari tabs：信息密度或相关性偏低

    高频关键词

    code · product · model · models · coding · use · time · day · anthropic · qwen · claude · team

    我的学习趋势

    今天的高信号内容主要集中在 AI。建议继续把阅读沉淀到可执行项目、求职表达或语言学习材料里，而不是只收藏链接。

    今天只做一件事

    Action

    读完 Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding，并写下 3 行可执行笔记：它说明了什么、和我有什么关系、下一步做什么。