Adam Dunkels

In previous posts we explored what happens if we treat LLMs as processors that run Markdown as their machine code: a user-space IP stack written in Markdown and a BASIC interpreter written in Markdown. Today we are going to look at an emulator for a 6502

How AI coding agents built a city-scale IoT asset tracking simulator in weeks, testing 100k+ Bluetooth assets across Birmingham before hardware arrived.

If we view Claude Code as a processor and Markdown as its machine code, what can we do with it? Today, let's build a BASIC interpreter.

Here is a fun thought: if we ask Claude Code to act as a user space IP stack, how quickly will it be able to respond to a ping?

AI coding agents build a working Matter smart home device with Thread networking, embedded firmware, and an Android commissioning app.

Most product ideas fail because nobody needed them in the first place. And by the time we figured that out, it may be too late. But there is a simple question that can cut through that particular fog: Whose problem are we solving?

An IoT device can pass its hardware watchdog check while doing nothing useful. Multi-layered watchdogs catch what hardware alone cannot.

We built a testbed with 100+ Bluetooth beacons on nRF52840 dongles to stress-test an IoT system at scale, catching firmware and OTA bugs before customers do.

This was a ridiculous idea I had one day: wouldn't it be fun to have a personal, portable laugh track - you know, like in those old sitcoms where there would be a canned laughter after every joke. So I figured it would be a fun project to. This is the res