AI and VoIP Blog

I set up a local AI coding agent on my Mac with Pi, Ollama, and Gemma 4.

A deep dive into how LLM tool calling actually works under the hood — from JSON schemas to grammar constraints to raw token generation.

Learn how to run Gemma 4 on Raspberry Pi 4B using llama.cpp.

Build a Raspberry Pi web page watcher that sends phone notifications when content changes.

Step-by-step guide to making and receiving SIP to PSTN calls using Twilio and Zoiper.

Interactive WebGL 3D experiment rendering a spinning cube in your browser.

Learn how to get reliable structured JSON responses from LLMs using Ollama's JSON mode and schema enforcement for production-ready AI applications.

Build a live voice AI agent using OpenAI's Realtime API and Twilio SIP trunking.

Build and install SIPp from source on Linux with SSL, PCAP, and SCTP support.

Run KittenTTS locally with a Gradio web interface for high-quality text-to-speech.