Orionfold · an AI studio

Grow 10x with AI, fast.

Fast, easy AI journey with expert playbooks, open software, custom models, on personal devices.

Get the books Sponsor the work

Most AI sends your data to the cloud.Ours runs on your own machine, open and private.

AI Native Platform software cortex software fieldkit software Arena software Kepler model neosignal software AI Native Business book AI Research on DGX Spark book Patent Strategist model

14 apps · 8 models · 3 books · all built here

A small gold desktop computer glows on a plain desk at night, sending bright beams of light up into a wide sky full of stars and constellations.

Built on the NVIDIA stack

Real AI research, run on one desk.

Our reference machine is the NVIDIA DGX Spark, a small AI supercomputer that fits on a desk. The book, the open models, and the tools were all proven on it first, with no cloud. See the whole body of work gathered in one place.

See the NVIDIA-stack work
Why now

AI is moving fast. The gap is growing.

Every few months, AI gets better, faster, and cheaper. People who build with AI pull ahead. People who wait fall behind. The difference isn't talent or effort. It's whether your work is built to run on AI.

New tools every week

AI changes so fast that a once-a-year plan can't keep up.

Skill, not staff

A 10x worker isn't busier. AI helpers run their daily tasks, right on their own computer.

Open tools caught up

Open AI now matches the big paid tools. No lock to one company. No fee per person.

Behind the scenes
Open vs closed, in tech terms. Open models are yours to download and run on your own machine. Closed models live behind a company's paid service. On public benchmarks (shared tests that score AI on the same questions), the best open models now land within a few points of the closed ones.
What to do

Own your AI setup.

You don't need a bigger team or budget. You need four things you control.

01
Expert playbooks

Books, notes, and real stories that turn hard lessons into steps you can follow.

02
Open software

Run AI helpers and tasks yourself. Open, with no cap on what you automate.

03
Custom models

AI trained for your field, like patents, security, law, money, or health. Not a one-size chatbot.

04
Personal devices

A computer strong enough to run that AI in private, right on your desk.

Behind the scenes
Open software means the code is public and yours to run, with no fee per user. Custom models start from open base models like DeepSeek-R1 and Qwen, then get fine-tuned (trained a bit more) for one field and ship as GGUF files (a format that runs on your own GPU). Personal devices means Spark-class hardware: a small desktop, around $3,000, built to run these models offline.
How we help

Orionfold builds all four.

We're the studio behind a full AI kit. Open software, custom models, and the guides to run it all in private. Every piece links from one place.

Runs on your own computer Private by default Open source Custom models for your field Tracks models, chips, and clouds Automate daily tasks Drive it all with code
Behind the scenes
The AI Native platform runs on your own machine. neosignal tracks how AI models, chips, and clouds stack up. fieldkit is a Python toolbox of patterns proven on Spark. The AI Native API lets you drive it all with code. Any AI feature here runs on a small server function, so keys and private data never reach your browser.
AI Research on NVIDIA DGX Spark book cover
The book

AI Research on NVIDIA DGX Spark

Real notes from doing AI research on one desktop. The NVIDIA DGX Spark packs huge power into a small machine, so you can push local AI further with no cloud needed. Every lesson is backed by code that runs.

Local AI NVIDIA DGX Spark Petascale desktop Backed by code
Explore the book
Orionfold Arena poster art
The software

Orionfold Arena

One screen to run, compare, and score the AI models on your own desktop. Watch live speed and memory, rank models on a private leaderboard, and test two side by side. Nothing you type leaves your machine.

Runs local Leaderboard Live telemetry Private
Explore Arena
Patent Strategist poster art: Prior Art
The models

Patent Strategist

An AI built to reason about patents (legal cover for inventions). It runs fully offline on a small desktop, so your ideas and filings never leave the room.

Patents Runs offline Open weights
Explore Patent Strategist
Story

From the build log.

All stories →
Three-panel comic. A builder pins a card to a big public roadmap board with Shipped, Building now, and Planned lanes, a diverse community cheers and raises coins to vote, and a lane lights up Funded under a blue constellation. A banner reads: Build in public, vote with dollars.
Building in public

Building in public

Every story here, every number and honest miss, I put out in public on purpose. This is why I build in the open, and how you can help shape what comes next.
Three-panel painted comic. A builder's finger hovers over a glowing key while a screen warns of a six hour job, then he studies three small glowing gauges on his workbench, then he relaxes as a small gold AI computer runs the job as a pure test, with a blue star constellation in the window. A banner reads Measure first, spend second.
Models

The run I almost wasted

I built Kepler, a space-math model, on one desktop. Three cheap checks told me the big training run would change nothing. I ran it anyway, and proving them right was the win.
Spotlight comic. In a cozy night reading nook, an open glowing book pours friendly code into the air that becomes a blue star constellation and a tiny running app, with a small gold desktop computer glowing beside it. A banner reads: Read it, run it yours.
Books

A book you can read and run

I write books you can run, not just read. The code works and every page is online. Here is why open, reproducible books beat theory behind a paywall.