有人嫌 ScreenStudio 导出慢、中文字幕差，直接拿开源的 Cap 改了个自用版，思路挺值得参考的。
博主订阅 ScreenStudio 近一年，遇到导出速度慢、中文字幕识别转写弱、无法调整倍速等问题。作者选择开源项目 Cap 作为基础进行改造，按自身需求定制功能。改造完成后决定不再续费 ScreenStudio 的订阅。
ScreenStudio 号称最强录屏软件快订阅一年了。 发现确实挺好用，但也发现不理想的地方，比如导出很慢，中文字幕识别转写很弱，没法调倍速等。 既然 AI 这么强了，不如自己写一个自用。 发现开源中最近接的是 Cap，地址我放评论区。 拿过来按自己需求改造，ScreenStudio 以后就不续费了。 💬 4 🔄 0 ❤️ 4 👀 1337 📊 5 ⚡
OpenRouter 出了个 cookbook，教你把文档丢给 agent，让它自己学会调 OpenRouter 的接口，搞自动化的可以看看。
Figma 财务团队用 AI 复刻竞品来验证收购标的值不值钱，还造了 AI 版高管来审备忘录，这套打法挺新鲜的。
11月10日在SF，FactoryAI的David Roster手把手带你用真实项目练怎么安全把活儿丢给agents，想学 delegates 流程的可以看。
The non-obvious part is the fusion design: instead of relying on a vision-language model alone, OCR readability features feed a CatBoost classifier alongside it, with SHAP explanations attached to every decision. Teams building content-moderation or quality-scoring pipelines can copy this rendering-plus-ML pattern on SageMaker. It also shows how Amazon turns model outputs into seller-facing guidance rather than just internal metrics.
教你不改模型就能让 Haiku 干 Opus 5.5 的活：59 秒替代 7 分钟，token 省一半，思路挺实用。
The core implication is that official benchmark scores are increasingly unreliable signals for model selection, and community-run alternative evals are becoming the de facto quality check. Builders choosing models for production should cross-check lab claims against independent altbenches before committing to an API or fine-tuning pipeline. Following Artificial Analysis's rebuilds offers a practical sniff test without waiting for lab corrections.
