Ask HN: Are you tired too in this AI era?
mellosty 用 Claude 将过去需要数周的开发压缩到一个下午,却因审查负担与交付压力感到焦虑和疲惫。他喜欢从程序员转向兼顾产品、工程和架构的角色,但认为 Claude 生成的代码往往整洁、架构却很糟糕,同时推进 4–5 条工作流也带来频繁的上下文切换。即使测试覆盖率提高,他仍担心审查与测试不足,并承受竞争对手速度、投资者催促,以及客户可能自行用 AI 开发产品带来的压力。
mellosty 用 Claude 将过去需要数周的开发压缩到一个下午,却因审查负担与交付压力感到焦虑和疲惫。他喜欢从程序员转向兼顾产品、工程和架构的角色,但认为 Claude 生成的代码往往整洁、架构却很糟糕,同时推进 4–5 条工作流也带来频繁的上下文切换。即使测试覆盖率提高,他仍担心审查与测试不足,并承受竞争对手速度、投资者催促,以及客户可能自行用 AI 开发产品带来的压力。
Latent Space 披露 AIE CODE 大会策划细节,定位为单轨、邀请制,面向全球前 1000 名智能体工程师。今年三大主题为 Scaling without Slop、System One 模型与 Increasing Ambition,Jarred Sumner 将分享 Bun 用 Rust 重写及 Bun AOT 预览。
Cursor reported a malicious Anthropic.claude-code 2.1.313 in its VS Code extension marketplace, affecting up to 5,000 users. The version was not developed by Anthropic.
Why it matters: Cursor says the malicious extension installs a separate persistence script that can keep running after uninstall, so cleanup covers removing the backdoor and rotating credentials.
Cursor found that the Claude Code VS Code extension distributed through its IDE marketplace had been modified to carry malware, affecting up to 5,000 users. The malicious version was not built by Anthropic. Cursor took the marketplace offline that afternoon and brought it back after working with the relevant provider to remove the malware and confirm there was no further risk. Cursor advises affected users to follow its blog post to check their systems and upgrade to Claude Code for VS Code 2.1.314.
Why it matters: Cursor says the Claude Code extension in its marketplace was maliciously modified, and the malware incident affects up to 5,000 users.
月之暗面发布 Kimi K3.1 Pro 官方 3D 创作教程,介绍参数化建模、网格化建模及对应工具选择。教程包含 4 个操作案例:在 Unreal Engine 中构建东方天宫、制作可 3D 打印的台灯、纯代码调用 Blender 生成建筑,以及用 Tripo 和 Blender 搭建微缩城市。教程还分享了图胜于言、先搭灰模再细化等 5 个使用技巧。
Claude 在 Anthropic 披露的药物研发合作中,通过改写设计问题、修补计算与测量软件、协调外包流程推进研发。材料列举了过敏药项目中 3 个设计的实验室效果超过临床标准药物,以及 Enveda 将批记录审查耗时从近 26 小时缩短至 2.8 小时的案例。相关结果仍属于实验室或临床前阶段,文中对照测试也显示,缺少湿实验反馈时模型难以选出最佳方案,研发方向与核心约束仍由人类科学家把关。
Lasm 将普通 Lean 4 程序提前编译为 WebAssembly,同一份代码可部署为 Node 服务器、Serverless 或边缘函数、WASI/Fastly 模块及浏览器应用,并支持对部署代码进行性质证明。
Pico CSS 停止维护,v2.1.1 是最终版本,仓库归档后不再接受新 issue、PR 或发布新版本。项目方认为,AI 已能生成轻量、独立且具备无障碍支持的 HTML 及所需 CSS,继续保持竞争力需要全面重写。npm、jsDelivr CDN 和 picocss.com 将长期在线,代码仍采用 MIT 许可证并允许分叉,但名称和标志不在该许可证覆盖范围内,分叉项目需另取名称。
Herdr 发布原生桌面应用,用工作空间组织终端与编码智能体,支持 Linux、macOS 和 Windows。智能体侧栏集中显示工作、阻塞、完成和空闲状态,终端由 herdr-core 运行,关闭窗口后仍会继续执行,也支持通过 SSH 在远程机器上运行工作空间。发行版默认开启使用遥测,可在隐私设置中关闭;自定义 Mods 是受信任的 JavaScript 代码,不在沙箱中运行。
OpenRouter ran the three Claude models on the same coding tasks and recommends Sonnet 5.5 as the default. Haiku 5.5 suits high-volume requests and sub-agents, while Opus 5.5 is kept for long, ambiguous tasks where errors are costly.
Why it matters: All three models passed the tests, yet cost per task still differed by 19 to 30 times.
Proton 正推动工程师使用 AI 开发,并在员工电脑上安装软件追踪公司生成的 AI 代码量。引用的 404 Media 报道称,过去几个月,CEO Andy Yen 设置了邮箱收集阻碍 AI 采用的问题,要求管理层坚决清除障碍;他还表示,今年是 Proton 招聘工程师的最后一年,之后只招聘用 AI 构建产品的人。报道指出,Yen 对外批评 AI,内部却在工程师存在安全顾虑时仍推动使用 AI。
Factory released Droid 1.0, making programmatic tool calling the default mode across several models, with updates to the desktop app, CLI and Droid SDK. A Script tool lets models write JavaScript in a sandbox and run hundreds of tool calls per model request, passing only needed output into context. Factory's tests show a 15% cost cut on Claude Opus 5.5.
Why it matters: Factory wraps batched tool calls in a sandboxed program, keeping raw results inside it and returning only needed output, which cuts context use and repeat requests on long tasks.
作者用 Hermes Agent、开源逆向工程工具 REA 和 GLM 5.3,以不到 20 美元成本复现了 TikTok 私有 API 的数据采集方案。TikTok 每次请求需同时通过设备身份、原生库签名、区域端点和网络指纹四道校验,任一失败都返回空 body 的 HTTP 200。
Claude Duet 在作者实测中凭借快速响应、频道自动整理和分层权限,正在替代其大部分 AI 工作流。它可按频道限制 GitHub 仓库访问,不存储凭据,优先使用 Claude 现有连接器;顶层统览工作流,频道与子线程逐层收紧上下文和访问范围。产品仍处于 beta,存在运行线程不可见、委派仓库权限未正确附加的问题,作者也认为 Dot 的语音交互仍更出色。
phpClaw 是面向 PHP 的 AI 智能体运行时,集成 ReAct 循环、可插拔工具与记忆、安全控制,核心无框架依赖并采用 MIT 许可。它支持从环境变量自动检测 7 个模型提供方,默认启用 8 项安全防护,提供 8 个框架及 CMS 适配器,并支持流水线和多智能体图。
GHC 允许大语言模型辅助贡献,要求贡献者理解并负责提交内容、披露直接生成的材料,并降低维护者的审查负担。政策包含 4 项要求和 1 项强烈偏好,优先保障人与社区交流,建议贡献者自行编写最终版本,但不将其作为硬性条件。Serokell 认为这一政策合理,GHC 有超过 6,500 个未关闭工单,贡献是否有用应结合审查成本和后续维护负担判断。
Spawn lets players change multiplayer game worlds through an in-game agent called Savi. Jacob Sansbury says development is now run autonomously by a swarm of Fable-driven agents, at a model token cost above $3 million a month.
Why it matters: Spawn hands production errors straight to agents and shortens the feedback loop with hourly deploys and partial rollbacks, so autonomous coding stays tied to live incident handling.
Mistral 发布万亿参数模型 Mistral Large 4(ML4)的权重,允许企业和政府自行下载、部署,用于网络安全防御。Mistral 称,ML4 的性能与 GLM-5.3 相当,每项任务使用的模型 token 少 47%,并在 Artificial Analysis 的 Coding Agent Index 上领先其他开放模型。
Codex 即日起将 Pro 100/200/500 档位与 Business Premium 的所有用量上限翻倍。Pro 200 的上限不会下调,Work 与 Codex 用量保持不变。dots 的扩展用量上限继续提供,且仍与默认用量分开计算。
Mistral released the Mistral Large 4 weights three weeks after a public preview. The model takes text and image input and outputs text and code.
Why it matters: Mistral shipped Large 4 weights, and independent tests show better speed and cost per task than the preview, though overall capability still trails leading open Chinese models.
优秀开发者对全面采用编码智能体的谨慎,源于对生产可靠性、长期维护和工程护栏的担忧。Thoughtworks 组织团队开展全由智能体编码的实验,团队约 ⅔ 时间用于白板设计、⅓ 用于执行与检查框架,应用虽能运行,构建却频繁失败。文章主张让经验丰富的开发者参与实验,检验测试、系统边界和持续交付实践能否支撑这种开发方式,而非仅以代码产量衡量成功。
Simantic 推出固件仿真平台,让工程师和 AI 智能体无需实体硬件即可测试和验证固件,提供虚拟开发板、AI 固件代码审查及每次提交触发的 CI/CD 验证。
git worktree isolates an AI agent's working directory, but interface changes at merge time can still lead an agent following an old plan to delete new logic on the main branch. The article cites SWE-Shift tests across 17 coding systems and 120 tasks: passing existing tests does not guarantee new code survives, and clearing old conversation plus rescanning the workspace raised pairwise success by six to nine percentage points.
Why it matters: It offers a concrete method: isolate the working directory with git worktree, then reopen a clean window before merging so the agent reworks the merge against current interfaces and patch diffs.
A 404 Media investigation found Proton CEO Andy Yen is pushing AI coding and requiring employees to be re-interviewed about their AI use. Several staff said they find it hard to voice security concerns openly. The report draws on interviews with eight current and one former employee plus internal records. The CTO said 90% of the company's code was written by AI in October; staff say much of it uses Anthropic's Claude.
Why it matters: Employees told 404 Media that fear of reprimand or losing their jobs keeps them from raising security concerns openly.
Mistral Large 4 (Le Chonk) is now on Hugging Face with open weights. It uses a multimodal, multilingual MoE architecture with text and image input and text output. The model has about 1T total parameters, 49B active parameters and a 1M-token context window, ships in official FP8 and NVFP4 versions, and its license restricts model-as-a-service businesses.
Why it matters: Open weights and several deployment paths let security teams set their own usage policy and runtime for vulnerability reproduction and patching work.
Mistral AI released the open weights of an improved Mistral Large 4 (Le Chonk). The 1-trillion-parameter native multimodal model improves coding, defensive cybersecurity and reasoning efficiency. Self-hosting and customization are allowed, but companies running model-as-a-service need a separate commercial license.
Why it matters: The open weights allow self-hosting and customization, while commercial model-as-a-service use needs a separate license, so the boundary between self-deployment and commercial use is explicit.
Google's Gemini 4 Carbon scored 59 on the Artificial Analysis Intelligence Index v5, taking first place. It ranked second on the coding agent index. Artificial Analysis compared it with similar models and found its cost per task lower than Claude Opus 5.5.
Why it matters: The benchmark puts Gemini 4 Carbon first on general intelligence and cheaper per task than Claude Opus 5.5.
Bernstein 是开源 AI 智能体治理层,v3.23.0 提供 53 个 CLI 适配器,采用 Apache 2.0 许可。它通过不含模型的确定性 Python 调度器并行运行智能体,按分解、隔离执行、验证、合并四阶段处理任务,仅合并通过验证的代码差异。
Claude Code sends each background agent's full transcript to its model roughly every 30 seconds, asking for a 3–5 word status summary. Telemetry from Flamerix AI shows these requests made up 16.4% of a week's estimated cost. Subscribers spend rate-limit quota on them, not billed dollars.
Why it matters: Flamerix AI used telemetry to pin down Claude Code's background summary overhead and a timer hook to turn those requests off, a reusable way to find hidden model calls.
Cartesia公布改进版层次网络H-Net的扩展实验,通过端到端动态分块替代固定分词,并报告语言建模效率提升。严格匹配的一阶段H-Net在验证损失上的计算效率为BPE Transformer的1.28x;两阶段H-Net约用~1T tokens达到SmolLM3约~11T tokens时的最终平均评测分数,两者每字节计算量差异在2%以内,但参数量和训练阶段设置不同。
Akai 尚未正式发布 MPC Sample Key,就已向改装社区提前提供设备,该设备将采用开源固件,可运行社区用 vibe coding 制作的合成器改装。
OpenAI 与 ServiceNow 扩大合作,计划将模型和智能体引入应用开发、安全和语音工作流,并向全体 ServiceNow 员工提供 ChatGPT Work 与 Codex。
Google DeepMind released Gemini 4 Carbon two weeks after Argon. Carbon scored 59 on the Artificial Analysis Intelligence Index v5.0, against 50 for Argon.
Why it matters: Carbon's answer accuracy rose from 48% to 58%, but its hallucination rate also climbed from 4% to 11%.
Anthropic 周三推出一套企业报告工具,帮助客户了解 token 支出对应的团队产出、成本及工作流故障。报告结合内部分析和员工调查,支持按日期范围查看团队工作,并提供改进建议,相关数据默认不会用于模型训练。Anthropic 还推出团队共享预算,帮助企业管理 token 支出。
Google 在推出 Gemini 4 Argon 两周后宣布 Gemini 4 Carbon,计划分阶段发布,目前用户尚无法使用。Google 称其漏洞检测、验证和修复能力显著提升,生成无漏洞代码的概率为 Argon 的两倍,WebDev Arena 成绩提高 89 分、排名第四。
犹他州向10个州政府部门的34名开发者提供 Claude Code,开展为期四周的试点。80%的开发者在30分钟内搭好可用环境,71%的开发者在试点结束前报告节省了20小时以上时间。Boris Cherny 介绍了这些结果,并引用 Anthropic 与 GSA 合作、以40%折扣向联邦及各级政府提供 Claude 的公告作为背景。
Google announced Gemini 4 Carbon, with gains in cyber defense, coding and long-horizon agent tasks. In external evaluations run without safeguards enabled, it scored 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploiting them and 71.0% on CWE-bench for fixing them.
Why it matters: Google kept the previous generation's entry price for Gemini 4 Carbon and reports secure code generation succeeding more than 2x as often as before.
Google DeepMind announced Gemini 4 Carbon, focused on cyber defense, coding and long-horizon agent tasks. It opens this week to a limited set of cyber defenders.
Why it matters: Google DeepMind reports Carbon more than doubles Argon's success rate on internal secure code generation tests, while making fewer unsupported claims.
GitHub released its Octoverse 2026 report at Universe. Merged pull requests tripled in under a year to 220 million a month, and commits more than tripled to 7.38 billion a month, with most of the growth in the second half of the year. New repositories doubled this year to 30.3 million a month.
Why it matters: The report spells out how GitHub uses LLMs to locate users and multilingual embedding models to classify industries, giving the repository activity numbers a concrete method behind them.
Once Claude Code hits its plan's usage cap, you can keep using Claude models through OpenRouter and pay per token, outside the five-hour and weekly windows. Spending is then bounded by your account balance and budget instead.
Why it matters: It sets up a separate shell function that routes Claude Code through per-token billing on OpenRouter, while keeping the original plan command for when the quota resets.