Engineering case study · Full-stack

Agrolink

A farmer-to-buyer marketplace prototype with crop discovery, dashboard views and authentication/profile integration.

Marketplace prototypeSource inspected · Independent project
Repository (opens in a new tab)

The problem

Explore a simpler way for buyers to discover produce and for farmers to present their listings.

The implementation

A React marketplace provides search, crop, district and price filters over sample data, with separate farmer/buyer views and an authentication layer.

System structure

Architecture in practice.

  1. React Router separates public marketplace pages and guarded farmer/buyer views.

  2. Client filter state derives the visible produce list from locally defined sample records.

  3. Supabase Auth and profile-query wiring are present in this external project's source.

  4. Dashboard statistics, saved products and activity remain demonstration data.

Engineering decisions

  • Keep crop filtering on the client for the small sample catalogue.
  • Use shared product cards and a contact dialog across marketplace views.
  • Model authentication/profile state separately from browsing filters.

Challenges & tradeoffs

  • Category, district, price and search controls need to combine predictably.
  • A prototype must distinguish real authentication wiring from sample catalogue and dashboard data.

What the project teaches

  • UI completeness does not establish end-to-end marketplace persistence.
  • Reusable filters and product components can connect public and dashboard experiences.
  • Source and contribution history help establish the actual scope of an AI-assisted prototype.

Current scope & limits

  • The inspected farmer 'List Produce' action shows a toast and closes the dialog; persistent produce creation is not implemented there.
  • No verified transactions, payments, farmer income improvement or live-user metrics.
  • The README-listed demo returned HTTP 403 in this audit, so a functioning deployment was not independently verified.
  • This external project uses Supabase; the portfolio itself does not require Supabase or Lovable.

Contribution & source evidence

Independent AI-assisted prototype. Repository history records Lovable-assisted implementation and owner documentation updates.

The case study reflects inspected source. External project test suites and business outcomes were not independently evaluated.

Let's talk about practical engineering.

Contact Zahid