即使在2026年，把AI从demo做成产品也不是一件容易的事。比如中翻英，大家都觉得早就被LLM解决了，不就是调个API的事情嘛。但我们最近因为要把Superlinear Academy社区加入一个中翻英自动同步的功能，才发现开发体验这么差。
这个问 …
From Process Certainty to Outcome Certainty: A Different Kind of Confidence in the Age of AI
Even in 2026, turning an AI demo into a production-ready product is surprisingly hard. Take Chinese-to-English translation. Everyone assumes LLMs solved this ages ago—just call an API, right? But when we recently tried to add an automatic translation sync feature to the Superlinear Academy community, we discovered just how …
使用AI生图助力天文学观测摄影
一切都缘起于深圳市天文台德铼老师给我的一段日面摄影的视频文件。他们的设备因为一些原因，CMOS积累了很多灰尘，同时在观测太阳时 …
Leveraging Generative AI for Astronomical Image Calibration
It all started with a video file of the solar surface shared by Mr. De Lai from the Shenzhen Astronomical Observatory. Due to various environmental factors, their CMOS sensor had accumulated significant dust. When observing the sun with a high f/40 focal ratio telescope, these dust specks became painfully …
我的2025：一个凡人的修仙旅程
我从五六年前开始，就会记录每个小时花在了什么事情上。积累了五六年的数据以后，去年我第一次把这些数据扔给大哥 …