Everything, in one place.
№ 010 · Case StudiesFrom Subscription to Self-Hosted: A Studio Case Study
Not the mapping of which tool replaced which, but the story of the move itself: why we decided to leave…
№ 009 · Behind-the-scenesA Week of Coding With a Local Agent, No Cloud
We spent a working week coding with OpenCode pointed at a local Ollama model and nothing else. No cloud agent,…
№ 008 · Builds & ToolsThe Cheapest Mac That Runs Serious Local AI
We get asked which Mac to buy for local AI more than any other question. The honest answer is not…
№ 007 · Builds & ToolsLocal Audio Cleanup: Demucs and WhisperX
A build log for the studio's interview chain: Demucs to isolate the voice out of a noisy field recording, WhisperX…
№ 006 · Builds & ToolsA Fully Local Image Pipeline With ComfyUI
How the studio built a repeatable image pipeline that runs entirely on one Mac: a node graph instead of a…
№ 005 · Behind-the-scenesThe Open Source AI Atlas: How We Evaluate Local Tools
A look inside the Atlas, the studio's running record of every local AI tool we test. The same template for…
№ 004 · Case StudiesTesting OpenJarvis End-to-End: What Worked, What Didn’t
Our working test notes on OpenJarvis, published in the open. This is the scaffold of a hands-on review with the…
№ 003 · Case StudiesWe Replaced Our Paid AI Tools With Open-Source Local Ones
A real swap, not a thought experiment. We took the paid AI tools the studio was leaning on and replaced…
№ 002 · Builds & ToolsWiring OpenCode to a Local Model: A Build Walkthrough
A build log, not a brochure. We wired OpenCode to a model on our own machine and watched it edit…
№ 001 · Builds & ToolsHow We Built Our Local AI Stack
The exact stack the studio runs every day: compute, model, tools. What we chose, what we rejected, and what we…