Stories by deadprogram

Tinyflight: Real ESP32S3 drone firmware can fly in the air and the browser too

Yzma now over 2.6X faster at embeddings than Ollama or Docker Model Runner

Jeyzma is TypeSafeAI minus $850M plus Go/TinyGo/llama.cpp in the browser

Jev in your local browser in Go with WebGPU using Jeyzma

100% local chat with a model in the browser using WebAssembly written in TinyGo

AI Takeover at GopherCon 2026

"How do you plan to dismantle the servers of your rival AIs?" [video]

Talking robot heads and one human with a tinfoil helmet [video]

Talking Heads from the Year 2053 – Episode One

Free Yourself from the Copilot Tax

Conversations With Our Future Robot Overlords (and their pet human) [video]

Talking Heads from the Year 2053 – Episode Zero [video]

TinyGo 0.41 on ESP32 – At Long Last, Wireless

Espflash – Go CLI/library for flashing ESP8266/ESP32 with no dependencies

Running local LLMs and VLMs on the Arduino UNO Q with yzma

Yzma 1.0 is out Go programs that run local models with hardware acceleration

Go apps can directly integrate llama.cpp for HW accelerated local inference

"Captions With Attitude" in the browser from local VLM using llama.cpp in Go

Pure Go hardware accelerated local inference on VLMs using llama.cpp

That Machine Always Lies

1 2