Applied AI studio
AI that ships.
We turn slow, manual work into AI products. In weeks, not quarters.
Work
Shipped, measured.
Enterprise work anonymized.
- 01MyColoringBook.ioAI SaaS · Next.js · Supabase · StripeAI coloring-page generator, idea to launch.↗
- 02Real-time social signalsGlobal media company · GCP · GeminiLive posts classified, editors alerted in Slack.↗
- 03Multimodal content searchDigital publisher · Cloud Run · FirestoreA whole archive, searchable by meaning.↗
- 04Legal review assistantLaw firm · RAG · extraction85% less document review time.
- 05Commerce rebuildWordPress · PHP · Core Web Vitals9s to 1s load, 47% more conversions.
- 06RUBY: RewindBrowser games · parallel AI agents37 original games and a roguelite, in days.↗
Services
What we build.
The model is rarely the product. We build everything around it.
AI products →
Assistants, RAG, SQL, voice and vision.
Agents & automation →
Agents and MCP servers, with human sign-off.
Rapid prototypes →
Working, on your real data, in 2–3 weeks.
Full-stack product
Next.js, TypeScript, Python, Postgres, cloud.
AI enablement →
Workshops that get teams building.
Specialist builds
Games, extensions, computer vision, Web3.
How we build
Agents run. Engineers ship.
Parallel coding agents clear the busywork. Our engineers own every line.
- 01Map
Find the friction.
- 02Prototype
On real data, in days.
- 03Prove
Beat the old way.
- 04Ship
Evals, guardrails, adoption.
FAQ
Questions, answered.
+−How do engagements work?
Most start with a two-to-three-week Prototype Sprint: a working prototype on your real data and a build plan. If it works, we build it, scoped to the project, or stay on as an embedded team.
+−What does it cost?
A Prototype Sprint starts at $20k. Builds are scoped and priced to fit the project. An embedded team starts at $25k a month. After a 30-minute call we send a fixed-scope proposal.
+−How fast can you ship?
A prototype is usually in real users’ hands in two to three weeks. Parallel coding agents compress the timeline; our engineers stay accountable for architecture, review and quality.
+−Which models do you use?
Whatever fits: OpenAI, Anthropic Claude, Google Gemini, open-weight or self-hosted models. Systems are built so models can be swapped without a rewrite.
+−Can you work inside an enterprise?
Yes: SSO, private cloud on GCP and AWS, security and legal review. Data stays in your cloud, never trains a model, and you own all code, prompts and evals. NDAs welcome.
+−What if AI isn’t the answer?
We’ll tell you. Sometimes plain software or a process change wins, and we’d rather say so before you spend money.
+−How is this site made?
By hand: plain HTML, CSS and JavaScript, no framework, no trackers. Every mark is one of thousands of square pixels moved by a small WebGL engine, in typefaces we drew ourselves. The real text stays in the page for search, assistants and screen readers.
Start
Tell us about the workflow.
What’s slow, manual or expensive?
- 01Reply in one business day.
- 0230-minute call.
- 03Fixed-scope proposal.
Received.
An engineer will reply within one business day.