Cosmetocare
Built during a paid internship, this project combined a modern web product, production-grade backend APIs, and applied AI models for cosmetic ingredient risk workflows. The full system covers user dashboards, admin validation, history/reporting, and structured data flows for ingredient-level and multi-compound analysis.
About This Project
Connecting cosmetic risk analysis, role-based workflows, and large ingredient datasets into one reliable product without fragile one-off scripts.
Delivered a unified architecture with a Next.js frontend, FastAPI services, Postgres/pgvector data layer, and deterministic + model-assisted pipelines for hazard and similarity analysis.
Shipped a complete private platform covering front-end UX, backend APIs, AI analysis paths, admin validation flows, and maintainable data operations.
Full-stack & AI Developer
2026
Private
Client
Technology Stack
Project Story
Connecting cosmetic risk analysis, role-based workflows, and large ingredient datasets into one reliable product without fragile one-off scripts.
Delivered a unified architecture with a Next.js frontend, FastAPI services, Postgres/pgvector data layer, and deterministic + model-assisted pipelines for hazard and similarity analysis.
Shipped a complete private platform covering front-end UX, backend APIs, AI analysis paths, admin validation flows, and maintainable data operations.
Insights & Takeaways
Highlights
- Internship delivery with production-style requirements and iterative stakeholder feedback.
- Unified frontend, backend, AI, and data-validation lifecycle in one platform.
Challenges
- Balancing prediction quality, traceability, and UX clarity for non-technical users.
- Keeping ingestion and analysis pipelines dependable while preserving admin control over approvals.
Lessons Learned
- Reliable AI products in regulated-like contexts require strong operational tooling as much as model work.
Related Work
Case Study
Auto Shutdown
Rust + React desktop app to automate PC shutdowns by timer, schedule or idle state.
Case Study
Azkari
Lightweight desktop app that displays Azkar reminders based on time of day — built with Tauri, SolidJS, TypeScript and Rust.
Case Study
Chef Kit
Modern multilingual cooking app offering recipe discovery, chef profiles, inventory management, favourites and push notifications — built with Flutter, Dart and Flask.
