Generative AI & Technology

Where AI meets the real world.

Live AI products built end to end, and in-depth writing on generative AI, large language models, and the technology reshaping how we build things.

Live projects

What I’m shipping

Live AI products, all running on the same infra. Click through to visit.

01 AI video production studio Showspring A complete agentic video production studio: 24 steps from idea to multi-channel publish. Show-bible-driven, so characters, voices, formats, and running jokes persist across episodes. Node · SQLite · Multi-LLM · Veo 3.1 / Kling 3.0 / Seedance 2.0 · ElevenLabs showspring.com 02 The show it produces The Doodle Cast Rusty and Oreo host the first AI-hosted podcast with doodles. Episodes and shorts are produced end-to-end through Showspring, and viewers can pitch their own dog into the cast. Showspring pipeline · Nano Banana 2/Pro · Veo 3.1 / Kling 3.0 / Seedance 2.0 · ElevenLabs thedoodlecast.com 03 Talk to the cast Live Character Voice Live voice conversations with the show’s characters, right on thedoodlecast.com. Personas come from the same show bibles and voice IDs that produce the episodes, so the character you watch is the character you talk to. SpeechRecognition · local LLM on RTX 5090 · ElevenLabs thedoodlecast.com 04 Multi-model deliberation Quorum + Bulb Two products on one shared debate engine. Quorum: Claude, Gemini, and Grok plus a local model debate in true sequence, then a Chair model delivers the verdict. Bulb is the gated idea-lab sibling. Node · shared debate engine · Claude / Gemini / Grok · local model council.overdigital.ai 05 Nightly AI forecasts WillWin Multi-event forecasting across World Cup, Eurovision, F1, Tour de France, and the Oscars. Predictions regenerate each night on a local Gemma model. Astro · llama-swap gemma4:26b · Brave · Cron willwin.ai 06 Generative ad creative · exploration Create Studio Brief to storyboard to generative video to scored creative output. An exploration of one-person ad pipelines; the production-grade version of these ideas lives in Showspring. Node · Nano Banana 2/Pro · Veo 3.1 / Kling 3.0 · multi-LLM scoring hapalops.com
AI Engineering — 10 posts
Deep Dive: Building the DaVinci Resolve Integration for Showspring
AI Engineering

Deep Dive: Building the DaVinci Resolve Integration for Showspring

Showspring renders broadcast-quality MP4s end-to-end on a server. So why ship them into DaVinci Resolve at all? Because some shows need a human’s last word — and getting a server timeline into the most opinionated NLE on the market took a layered integration design, an unsolicited Defender false-positive, and a tour through Resolve’s most-undocumented APIs.

Studio 8H in a Database — How the Doodle Cast Audience Got Built
AI Engineering

Studio 8H in a Database — How the Doodle Cast Audience Got Built

News-desk comedy needs an audience. Sitcom laugh tracks were wrong. Dense festival crowds were wrong. Weekend Update was right — and getting there took nine reaction types, a multi-take stem library, a density dial, a multi-model LLM planner, and one weirdly specific trick to make the goodbye bleed into the outro.

WillWin: predicting five championships nightly with one local LLM
AI Engineering

WillWin: predicting five championships nightly with one local LLM

WillWin is a personal technical experiment — a single open-weights 32B model running on a local RTX 5090 reads public data sources and publishes nightly probability estimates for the 2026 World Cup, Eurovision, F1, Tour de France, and the Oscars. Not a prediction service, not betting advice.

I built a voice chat with two talking dogs
AI Engineering

I built a voice chat with two talking dogs

A real-time voice chat on thedoodlecast.com lets you speak to Rusty and Oreo, and hear them talk back in character. Here's how it works, the stack behind it, and how to try it.

Introducing Bulb — Your AI Startup Idea Lab
AI Engineering

Introducing Bulb — Your AI Startup Idea Lab

Got a startup idea? Bulb puts it through the gauntlet: four AI evaluators with wildly different perspectives tear it apart, debate each other, then a final verdict tells you whether to build it or bin it. Free to use, no signup required.

How We Built the TheDoodleCast AI Chatbot
AI Engineering

How We Built the TheDoodleCast AI Chatbot

How we built a real-time AI chatbot for TheDoodleCast using Ollama, a local RTX 5090, Cloudflare Tunnel, and a custom streaming UI — without paying per token.