Skip to main content

Lab Notebook · Vertical SaaS

A solar-farm SaaS on the most modern stack in the portfolio — council-reviewed UX

Since March 2026 we've been building a multi-tenant SaaS for evaluating, financing, and managing solar-energy projects — site screening, financial analysis, AI-powered reports, 3D visualisation — on the newest major-version stack anywhere in the portfolio: React 19, Express 5, Vite 6. Nearly complete, not yet shipped, at the time of the audit.

JR
Jon RossFounder, vocabotics — 15 years building safety-critical systemsLab report · dated 1 March 2026
Verified by a human. Drafted with AI, verified by a human. Jon Ross, 1 Mar 2026
Living document. Reviewed 1 Mar 2026
Entry date
1 March 2026
Category
Tooling
Lead over the world
keeping pace, not a lead
Access
🔓 Public
ResearchPartly proven

Geospatial mapping, financial modelling, and 3D rendering, in one product, on the newest majors of every framework it touches.

React 19
Express 5, Vite 6, Tailwind CSS 4 — current majors, not legacy versions
3
disciplines in one product: geospatial mapping, financial modelling, 3D rendering
Council-reviewed
the council+gate method applied to product UX decisions

Honest evaluation

Partly proven

The stack kept pace with current major versions across a genuinely multi-disciplinary product; the platform itself was nearly complete but not shipped or live at the time of the audit.

What would prove or disprove it further

What would move this to proven: shipping the platform and getting real assessors or financiers to run a live evaluation through the geospatial, financial, and 3D workflow — the review process this platform used can catch design mistakes, but only real use tests whether the underlying product judgment was right.

The evidence — full reasoning behind the verdict

Verdict: partly-proven. Two claims run through this report. The first — that the AI-assisted approach can keep pace with a fast-moving ecosystem rather than lagging behind out of caution — is well supported: React 19, Express 5, and Vite 6 are current major versions, not legacy ones, used across a genuinely three-discipline product (geospatial, financial, 3D). The second — that the platform itself is a working, validated product — is not yet proven, because it wasn't shipped.

The report states its own limit plainly: "nearly complete" is a status, not a shipped claim. Council-reviewed UX is real process discipline that catches design problems before a user sees them, but a council review is not the same evidence as a real user actually using the product — and no real-user evidence exists yet here.

Most projects in this portfolio settle on a stack early and stay on it. This one asked a different question: could the AI-assisted approach keep pace with a genuinely current, fast-moving ecosystem, rather than freezing on whatever was current when the project started?

What it was

A multi-tenant SaaS for evaluating, financing, and managing solar-energy projects: site screening, financial analysis, AI-powered reporting, and 3D visualisation of proposed installations, aimed at the teams who assess and finance solar developments.

What we built

Research

A working platform on the newest major-version stack anywhere in the portfolio: React 19, Express 5, and Vite 6, styled with Tailwind CSS 4, with tRPC for typed API calls, MapLibre GL and deck.gl for geospatial visualisation, and React Three Fiber for the 3D site views — all running on Node.js 22, PostgreSQL 16 with PostGIS, and Redis, inside a pnpm/Turbo monorepo. UX decisions went through the same council+gate review discipline used elsewhere in the portfolio for architecture — applied here to product design, not just code.

What we learned — including the honest negative(s)

The real proof this project offers isn't in the solar domain itself — it's in the stack. Building on React 19, Express 5, and Vite 6 (all recent-at-the-time major versions) while also scaling to a genuinely multi-disciplinary product — geospatial mapping, financial modelling, and 3D rendering, together — is real evidence that the AI-assisted development approach can keep pace with a moving ecosystem, rather than lagging a version or two behind out of caution.

The honest edge, stated as the project's own words: "nearly complete" is a status claim, not a shipped claim. This platform was not live or shipped at the time of the audit, and this report doesn't round that up. Council-reviewed UX is a genuine process discipline — it catches design problems before they reach a user — but it is not itself evidence of real-user validation, which only happens once something ships.

Where it went / status

Active development, nearly complete at the time of the audit. Not yet shipped or live.

What is still open — kept visible

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

  • "Nearly complete" is the project's own status at the time of the audit — not yet shipped or live.
  • Council-reviewed UX is a real process discipline, not a substitute for real user validation once live.

Where this connects

Sources

  1. vocabotics project audit — Solar Farm Platform (React 19/Express 5/Vite 6, council-reviewed UX, nearly complete), Mar 2026vocabotics internal project history · as of March 2026

    We use cookies.