ESP32 512kB RAM with Tailscale, English to Python LLM and 8 containers local (opens in new tab)
The chip serves a web agent IDE, a native MCP server and a REST API, runs the code in a sandbox (with live web-request / fetch support), and drives its own screen + LED — up to 8 parallel PySpell processes on that same half-megabyte of RAM. Like MicroPython, but two syntaxes, the parser never ships to the device, and English is the third syntax.
Read the original article