Online skin analysis tool — website, admin panel, backend

A standalone, monetisable online skin-analysis product had to be built from zero — website, user accounts and the full backend logic behind it.

The analysis engine had to be fast, accessible and credible at the same time: returning results a lay user can understand, in real time via AI processing.

The product needed an admin surface from which content, users, analyses and metrics are all managed in one place, independently.

Website: fast, mobile-optimised, conversion-focused public surface with a landing page, pricing and a built-in checkout flow.

Analysis engine: image-based AI processing returning skin type, condition and recommendations; the structured result is saved against the user's history.

User accounts: registration, login, subscription/credits management, analysis history and personalised recommendations under the user's own profile.

Admin panel: content management, user and subscription management, an overview of analyses, conversion and usage metrics, with role-based permissions.

Backend: a Supabase + PostgreSQL data model, edge functions for processing, auth, storage and row-level security protecting the data.

Operations: monitoring, error logging, an architecture ready for A/B testing and content updates from the admin panel without code.

A live, monetisable product with a complete frontend–backend–admin stack, delivered from one hand.

Users get real-time, understandable analyses and can return to their past results in their account.

Content, users and metrics are managed independently from the admin panel — without developer involvement.