Deterministic-first
Everything that can be computed becomes rule logic, templates, or browser/server compute.
Back to home
Methodology
How we build products that do not need AI to run.
Four phases. Deterministic-first by architectural commitment. AI does the heavy lifting at design time so the product can run as pure compute - same answer every time, no live model call, forever.
AI upstream
AI helps create the rules, tests, docs, and product surfaces. The user path does not depend on a live model call unless interpretation is genuinely required.
Ownership is the deliverable
The handover includes source code, brand assets, deploy keys, domain wiring, and operating notes.
The Four Phases
The workflow from raw idea to owned system.
- 01Spec before prompt
Strategy
Define the problem, the user, and the boundary of what we will and will not build. Write the rule we are going to ship, not the prompt that simulates one.
- 02Prototype judged by humans
Design
AI-assisted, human-judged. Stitch and Google AI Studio for high-fidelity prototypes; reviewed for hierarchy, accessibility, and brand fit before code starts.
- 03Agents execute explicit specs
Build
Claude Code and Codex write production code under explicit specs. Tests are part of definition-of-done, not TODO.
- 04Source and keys handed over
Deploy
Vercel for frontend, Railway for Python backends, Supabase for data and auth. Source code, brand assets, deploy keys, and operating notes are transferred.
Worked Example
The CRS Calculator, four hours, end-to-end.
IRCC publishes the Express Entry CRS scoring tables. They do not change. The calculator at /tools/crs-calculator is a faithful encoding of those tables in TypeScript.
- 0:00
Pulled the CRS tables straight from the IRCC PDF. Claude transcribed each scoring section into JSON; we cross-checked against three sample profiles.
- 0:45
Wrote the scoring functions: age, education, language (CLB), work experience, spouse adjustments. Each gets a unit test.
- 2:00
Built the React form. Sliders for age + CLB, dropdowns for education + experience, and a result chip that updates on every keystroke.
- 3:30
SEO meta + JSON-LD WebApplication schema. Sitemap entry. Plausible tracking on result computation.
- 4:00
Deployed. Runtime path: browser compute, no model call per calculator run. Even if a million people use it tomorrow, the bill stays the same.
The Cost Model
Deterministic vs. traditional, at scale.
Traditional AI product
$0.01 - $0.50
per interaction
Margins shrink with every user. AI bill grows with traffic.
MDS deterministic product
Near-zero
per interaction
AI builds the rules at design time. Runtime is pure compute.
We use AI to build sophisticated rule engines, templates, and computation systems. Then they run forever at near-zero cost. You own everything.
Stack & Tools
What runs where.
Note the split: every AI tool is in the build column. The runtime stack is conventional and boring on purpose - that is how the deterministic cost model holds.
Strategy + writing
Claude - ChatGPT - NotebookLM - Gemini Deep Research
Design
Stitch - Google AI Studio - Figma
Build
Claude Code - Codex - Antigravity (parallel agents)
Review + audit
Codex cold-eye audit - ChatGPT red team - GitHub Actions
Runtime
Next.js 16 - React 19 - Tailwind - Supabase - Vercel - Railway
Have a deterministic problem?
Bring it. We will scope it through a workflow diagnostic.