How explicit organizational state helps agent platforms avoid exception drift

There is more technical content available today than any human being could read in a thousand lifetimes. Every topic has a dozen YouTube videos, three Substack

The Simplified Chinese Internet ran on reward and punishment. LLM training feels uncomfortably familiar.

Google DeepMind’s Benoit Schillings on AI-generated code, software engineering, and his lecture at Infobip Shift 2026 in Zadar.

The ClickHouse query behind our app's first screen took 85 seconds. A partition key, a sorting key, one precomputed column, flat query templates, and a big update. It ended up more than 200x faster.

There are lighter and more reasonable alternatives to GraphQL.

Everyone is trying to speed up software development by building their solutions out of ready-made bricks like third-party components or libraries. But we tend to forget what problems they may...

An opposite view on .gitignore files.

Asking AI to build me custom static site generator using only tools available in most POSIX systems.

Naming is communication, and clear language entails clear thinking. Why the craving for consistency rules is a refusal to engage, and how sloppy terminology degrades a codebase for humans and models alike.

Heidegger thinks about tools like so: a tool is something (a thing that has being in the Heideggerian sense, not necessarily an object) that represents an extension of human capabilities in some way. A tool is thus any kind of thing that *lets you do something you wouldn't otherwise have been able to*. The important (for us) conceptual leap here is that when a tool works well, or isn't broken, it develops what Heidegger describes as a "ready-to-hand" quality: the tool fades into the background as a kind of human-tool gestalt of a human with the additional capability afforded by a tool forms. As a simple example of this, consider eating with a fork. When the fork is well-designed and not broken, *you don't explicitly have "I am using a fork" in your conscious mind while eating dinner*.

Counting tokens is better than nothing. It’s also not the discipline.

A contrarian view

For a decade, identity and access management meant one thing: governing the humans who log in. Employee joins, gets provisioned, gets a manager, gets a

There was a brief window where AI agent security felt like a future problem. Organizations deployed copilots, coding assistants, and autonomous workflows on

To keep pace with modern business, data strategy is shifting toward more autonomous real-time systems that deliver intelligence at the moment decisions are

There are so many great places to learn about out there on the internet, and one of those great places is the legendary Linux Kernel. If you stumble around s...

multilinear map on some combination of scalars, vectors, covectors, and tensors

programming language

With their new company, General Magic, the creators of the Macintosh aim to revolutionize computing - again.

When people in Silicon Valley recall General Magic, they generally remember it as a tragedy. Originally an internal project at Apple, General Magic proposed making personal communicators that would let individuals get messages, perform computing functions and take calls remotely. In other words, a smart phone. Then Apple CEO John Sculley spun [...]

AI makes projects with weak engineering culture fail much faster.

Interview essays on how writers write— their workflows, note-taking, and storytelling, distilled through my lens. Show your workflow.

real-time operating system

TRON was an ambitious project to say the least, and after a good start, it suddenly came to a screeching halt.

Dealing with AI coding and swinging from feeling on top of the world building things faster to crashing and feeling like everything is pointless