Daily Brief

A compromised release of litellm, a Python package pulling three and a half million downloads a day, sat live for three hours installing a credential harvester for API keys, SSH keys and crypto keys along with a backdoor for remote command execution. It was caught by pure luck: the malware had a bug that crashed Cursor, and a researcher went looking. Saoud Rizwan uses it to mark how far the trust model has fallen. Zig's code of conduct now bans AI from pull requests, issues and even comments, because the team values growing contributors over collecting contributions. curl is weighing the end o

In 1945 Vannevar Bush described document scanning, OCR, speech to text, hypertext, search engines, a head mounted camera, and voice interfaces, in a single essay, before any of it existed. Kwindla Hultman Kramer uses As We May Think to set up the uncomfortable part: he was a baby programmer in 1995 writing HTML by hand and web servers in C, as excited about web pages then as he is about agents now. Web pages turned out to be a primitive, not a destination. His argument is that agents are the web page of this era, and the thing worth building is the AI native software that comes after. The tou

The Claude Certified Architect exam hands you six production scenarios and picks four at random, and Frank Coyle walks through them backwards, leading with the anti pattern in each one. Knowing what not to do is what points you toward what to do, the same way the design patterns movement of the early 1990s came with a catalog of the moves that quietly ruin you. Scenario one is a customer support loop, and the anti pattern is calling the model, taking the response, and using it. What you want instead is to branch on the stop reason, because the model cannot execute a tool at all. It only hands

Wisedocs processes medical claims that arrive as PDFs over 10,000 pages long, some of them larger than video files, through a pipeline of ML models spread across ten repositories nobody enjoyed touching. Denys Linkov's team spent six months collapsing that into a monorepo, and this talk is an honest audit of whether they should have just waited for the models to get good enough to do it for them. The benchmark he keeps coming back to is a single refactor task. With o3 it took three hours of back and forth in Cursor and still shipped ten major mistakes. Rerun on newer models, Sonnet 4.6 needed