DataLegacy - Engineering Data Management Platform
Built for Suncor's Engineering department (internal platform)
A full-stack engineering database web app I designed and built end to end - it manages capital-project packages, equipment tags, serial numbers, and maintenance tasks across industrial plants, with role-based access, single sign-on, and a full audit trail.
DataLegacy is a production engineering-data platform I designed and built end to end for my company's Engineering department. It replaces scattered spreadsheets and shared network folders with one auditable system for managing capital-project packages, pressure-equipment tags, serial numbers, and maintenance tasks across industrial plants.
The problem
Reliability and integrity work on a large industrial site generates an enormous amount of record-keeping: work packages, pressure-equipment tags and serial numbers, fitness-for-service history, and tasks that live across corporate systems like SAP. Most of it lived in disconnected spreadsheets and folders, which made records slow to find, hard to trust, and impossible to audit. I built DataLegacy to bring that data into one place, with proper access control and a complete history of every change.
What it does
- A 4-stage package wizard to create and track engineering work packages (work number, MOC, plant, train, status), with Excel and PDF export.
- A hierarchical equipment browser - Plant to Train to Class to Tag to A-Number to Serial Number - with a unified quick search and full-text search across notes.
- Role-based dashboards for admins, managers, and engineers, plus task views that read maintenance notifications and change actions from the corporate data warehouse.
- Tools for validated bulk Excel import, a package record editor, and PDF utilities.
- An audit-log viewer showing the full create / update / delete history with before-and-after diffs.
How it is built
DataLegacy is a Next.js 16 App Router application (React 19, TypeScript) backed by Prisma and SQL Server. Heavily-read datasets use a layered, offline-first cache (in-memory plus IndexedDB) so the UI stays fast and keeps working when the network is flaky.
- Next.js 16 (App Router, Turbopack) + React 19 + TypeScript 5
- Prisma + SQL Server for application data; a separate read-only connection to the corporate data warehouse for SAP / Enablon task data
- Tailwind CSS 4, Framer Motion, TanStack Table, Recharts, and ReactFlow for the UI
- ExcelJS and jsPDF for exports; Dexie (IndexedDB) for the offline cache
- Deployed to Azure App Service through a CI/CD pipeline
Security and compliance
- Single sign-on through Microsoft Entra ID (Azure AD), using certificate-based client authentication instead of a shared secret.
- Role-based access control across five roles, enforced both at the routing layer and again inside every API handler.
- An append-only audit trail recording who changed what, before and after, with route, IP, and user-agent.
- Hardened security headers (CSP, HSTS, and more), with secrets held in a managed vault rather than in the app.
What I am proud of
The part I am most proud of is that this is a real, trusted system used by an engineering team - not a demo. It went through a full multi-major platform upgrade (Next 14 to 16, React 18 to 19, Tailwind 3 to 4, and a NextAuth v4 to Auth.js v5 migration including the certificate SSO flow), verified end to end, and it holds up to security review.
Building it also deepened exactly the domain I work in every day: pressure-equipment records, package workflows, and reliability data. It is the clearest proof of what I am building toward - engineering software written by an engineer who actually does the field work.