Skip to main content

Lab Notebook · Own Product Line

CraftWorkz — keeping a second product line deliberately separate from the first

An AI-driven platform for generating digital products with a personal-assistant agent and a modular toolbox — built as a deliberately separate product from CraftKit, the dev-platform product, to learn what clear product boundaries actually cost and buy you.

JR
Jon RossFounder, vocabotics — 15 years building safety-critical systemsLab report · dated 1 June 2025
Verified by a human. Drafted with AI, verified by a human. Jon Ross, 1 Jun 2025
Living document. Reviewed 1 Jun 2025
Entry date
1 June 2025
Category
Tooling
Access
🔓 Public
ResearchPartly proven

Two related AI product platforms, built in parallel, kept deliberately separate — the boundary was the lesson, not the code.

Frontend + core features
working stage reached
Dashboard, wizard, marketplace, office mgmt
modular toolbox scope

Honest evaluation

Partly proven

The product-boundary discipline held; the platform itself reached frontend/core-features stage, with backend and marketplace depth still catching up.

What would prove or disprove it further

What would extend this further: a working, demoable backend and marketplace at the same depth as the frontend, which would move the platform-completeness claim from partly-proven toward proven without touching the (already-settled) boundary-discipline claim.

The evidence — full reasoning behind the verdict

Verdict: partly-proven. Two different claims sit inside this report, and they land differently. The product-boundary discipline — keeping CraftWorkz genuinely separate from CraftKit despite both evolving in the same window — held up in practice, not just on paper. The platform-completeness claim is weaker: a working frontend with core features in place, but backend depth and the tools marketplace lagged behind at the point of the audit.

The evidence for the boundary claim is negative-space evidence: no blurred roadmap, no shared user base by accident, across two products built in parallel by the same small team — the harder failure mode to avoid, avoided. The evidence for the incompleteness is stated directly in the project's own status: frontend/core-features stage, not a finished platform.

Building two related AI platforms at once raises an obvious question: where does one product end and the other begin? CraftWorkz exists partly to answer that question in practice, not just on a roadmap slide.

What it was

CraftWorkz is an AI-driven platform for generating digital products — websites, apps, documents — built around a personal-assistant agent and a modular toolbox: a dashboard, a project wizard, a tools marketplace, and office-management features.

What we built

Research

A working frontend with the core features in place: React 18, TypeScript, and Vite, styled with shadcn/ui and Tailwind CSS, state managed with Zustand, backed by a Node.js/Express service and PostgreSQL. The personal-assistant agent and the modular toolbox concept both reached working form on the frontend; the deeper backend and marketplace layers were still developing behind it.

What we learned — including the honest negative

The real lesson wasn't a technique, it was a discipline: CraftWorkz was deliberately kept as a separate product line from CraftKit, our own dev-platform product being built in the same window. It would have been easy to let features and users blur across two related-but-different platforms — a developer tool and a product-generation tool for a different audience entirely. Keeping the boundary explicit, even while both were actively evolving, meant neither product's roadmap got pulled sideways by the other's users.

The honest edge: this is frontend and core-features stage, not a finished platform. The backend depth and the tools marketplace in particular lagged behind the frontend at the point of this audit — a real gap, not a rounding error.

Where it went / status

Working frontend with core features live; backend and marketplace depth still catching up. It stands as the record of a genuine product-boundary decision made under the pressure of building two related things in parallel, one of several places in the portfolio where the discipline mattered more than the code.

What is still open — kept visible

The honest edges, next to the wins. This is what turns 🔬 into 🟢 — honestly.

  • Frontend/core-features stage, not a finished platform — the backend and marketplace depth lagged the frontend.
  • A personal-assistant agent and modular toolbox were designed and partly built; not all planned modules reached the same depth.

Sources

  1. vocabotics project audit — CraftWorkz (AI product-generation platform), Jun 2025vocabotics internal project history · as of June 2025

    We use cookies.