Lech Kalinowski describes building a physical terminal to interact with an OpenClaw instance and a remotely hosted language modelA language model learns patterns in sequences so it can interpret context and generate likely continuations such as text, code or structured responses.. The device combines a small OLED display for dynamic text with electronic paper for a more persistent view.
He explains the microcontroller, power management and backend architecture, then discusses practical failures involving wiring, displays and noisy hardware components. The model runs on a separate machineAI compute infrastructure is the hardware, facilities, networks, power, cooling, storage, and software used to train and run AI models. rather than the small microcontroller.
A demonstration shows terminal commands and an AI-assisted text role-playing game with generated characters and worlds. Kalinowski presents the project as an experiment in a quiet, physical interface for working with language models.
Watch on YouTube
