mediacutlet writes about pocket-tank, a project featuring a 14-million-parameter LLM that manages a virtual aquarium on an ESP32-S3 microcontroller. Distilled from a much larger 26-billion-parameter teacher model into a compact 7.56 MB file, the "brain" operates entirely offline without any network connection. The system uses a three-layer architecture consisting of a physics/reflex layer, an LLM advisor for decision-making (such as feeding or socializing), and a progression layer to manage long-term growth and life events like fish births and aging.
- The model is distilled from gemma4:26b into the smaller student version.
- Decisions made by the LLM are implemented via a reflex layer running at 25–30 frames per second.
- It supports an "installer" that allows users to flash firmware directly through a web browser using Web Serial.
- The project includes a PC simulator and support for QEMU emulation of the ESP32 hardware.
- The model is distilled from gemma4:26b into the smaller student version.
- Decisions made by the LLM are implemented via a reflex layer running at 25–30 frames per second.
- It supports an "installer" that allows users to flash firmware directly through a web browser using Web Serial.
- The project includes a PC simulator and support for QEMU emulation of the ESP32 hardware.
This page provides instructions and tools to install or update the Pocket Tank application on supported hardware, specifically the Waveshare ESP32-S3-Touch-AMOLED-1.8 board. Users can perform standard installations, updates that preserve existing data, or a full wipe via an "Erase" function using compatible web browsers like Chrome or Edge.
- The app includes a 7.5 MB model for local processing; no external communication is required once installed.
- Updating the firmware preserves fish names, badges, sand dollars, and decorations.
- To reset the tank manually without this page: hold `BOOT` and tap the screen to confirm the wipe.
- Troubleshooting involves waking a sleeping device by firmly pressing the `PWR` button or putting it into bootloader mode using `BOOT`.
- The app includes a 7.5 MB model for local processing; no external communication is required once installed.
- Updating the firmware preserves fish names, badges, sand dollars, and decorations.
- To reset the tank manually without this page: hold `BOOT` and tap the screen to confirm the wipe.
- Troubleshooting involves waking a sleeping device by firmly pressing the `PWR` button or putting it into bootloader mode using `BOOT`.
Dominik Kundel demonstrated the versatility of the Codex app server by having theCodex AI agent autonomously integrate itself into the game DOOM.
- Codex modified the DOOM engine and game maps to create a functional, in-game terminal. This allows users to interact with the AI agent directly within the game world to perform coding or game-related tasks.
- The setup uses an Electron app with a fork of `doom-wasm`. The agent patched the game data (Freedoom) and implemented a custom C file (`codex_terminal.c`) so the terminal renders natively within the engine rather than as a simple UI overlay.
- Codex handled the entire end-to-end process—coding, testing, and verification—with minimal human intervention. It even used Playwright to "play" the game itself to verify that textures and logos rendered correctly from different angles.
The project serves as a "demo-driven" proof of concept to show that the Codex app server can be embedded into any software environment or workflow, from IDEs to video games.
- Codex modified the DOOM engine and game maps to create a functional, in-game terminal. This allows users to interact with the AI agent directly within the game world to perform coding or game-related tasks.
- The setup uses an Electron app with a fork of `doom-wasm`. The agent patched the game data (Freedoom) and implemented a custom C file (`codex_terminal.c`) so the terminal renders natively within the engine rather than as a simple UI overlay.
- Codex handled the entire end-to-end process—coding, testing, and verification—with minimal human intervention. It even used Playwright to "play" the game itself to verify that textures and logos rendered correctly from different angles.
The project serves as a "demo-driven" proof of concept to show that the Codex app server can be embedded into any software environment or workflow, from IDEs to video games.
PycoClaw brings full OpenClaw agent parity to embedded hardware — a MicroPython-powered AI agent that can run on a $5 microcontroller. It features one-click flashing, a full agent loop, hardware control, multi-channel chat, persistent memory, and ScriptOs skills.
PycoClaw is an open-source platform for running AI agents on microcontrollers. It brings OpenClaw workspace-compatible intelligence to embedded devices costing under $5. Built on MicroPython, it supports multi-provider LLM routing, multi-channel chat, tool calling, extensions, over-the-air updates, and battery operation.
This article details the process of running a personal AI assistant on a low-cost microcontroller. It covers the use of Ollama for running large language models (LLMs) locally and MimicLaw for optimizing the model for resource-constrained devices. The author shares their experience with porting and running the models, along with the challenges and solutions encountered.
A blog about Emacs, Rust, and low-level systems programming.
