•TypeScript/React SPA handling large, real-time datasets with custom D3.js visualisations and useReducer using discriminated union action types for compile-time state guarantees; ResizeObserver-driven responsive SVG rendering across all viewport sizes. Demonstrates the performance-first, type-safe engineering culture that aligns with Omnea’s stack philosophy.

•End-to-end production SaaS platform with analytics dashboards, API key management, tiered billing via Stripe, and a Node.js REST API on PostgreSQL/Prisma; Jest and Cypress test suites cover critical paths; deployed on Docker/AWS with GitHub Actions CI/CD. AI-assisted development tools integrated throughout the engineering workflow.

•Owned the entire product end to end from greenfield setup: TypeScript/React frontend, Node.js REST API backend, 7-model PostgreSQL schema with Prisma ORM, Redis caching, D3.js analytics layer built from scratch, production auth (NextAuth.js, JWT, bcrypt), Docker deployment on AWS, and GitHub Actions CI/CD. Demonstrates the full-stack ownership model Omnea operates on.
•Engineered a custom multi-source data normalisation pipeline handling inconsistently structured CSV imports from four financial institutions, with deduplication logic and keyword-based auto-categorisation; the same class of complex data transformation problem found in procurement and spend visibility platforms.
•Built customer-facing analytics views translating raw financial data into actionable insights: interactive D3.js spend breakdowns, income vs. expense bar charts across a 6-month window, net worth tracking, and a budgeting engine with configurable thresholds and a green/amber/red progress system; prioritised clarity and usability over raw data display throughout.