About

The project

Starva is a personal training analytics dashboard built on my real Strava data. I built it to fill in the gaps in my public portfolio — Next.js App Router, GraphQL, MongoDB, and a proper CI pipeline. The data is live: activities sync from Strava via OAuth, land in MongoDB Atlas, and get served through an Apollo GraphQL API that runs as a Next.js route handler.

The components are built as a reusable kit — purely prop-driven, with the charting library wrapped behind a stable API so it stays swappable. Every push runs lint, typecheck, build, and Chromatic visual regression in CI.

Stack

• Next.js 16 (App Router)
• Apollo Server v5
• Apollo Client v4
• GraphQL 16
• MongoDB + Mongoose
• Storybook 10
• Chromatic
• Recharts
• TypeScript
• Tailwind CSS v4
• Vitest
• GitHub Actions

Who I am

I'm a front-end engineer based in Pickering, Ontario with 5+ years building production web apps. Most of that time has been at Huckberry, where I introduced Storybook, set up Chromatic visual regression, co-led a design-system rebrand tied to Figma tokens, and have been the main reviewer of component work across the front end.

Starva is my way of proving out the back-end and infrastructure side of things in public — something a UI-focused role doesn't always give you room to do.