Welcome to Fragrance Engine

We built Fragrance Engine because every other fragrance app turns into a catalog or a marketplace. We wanted the opposite: a tool that helps you wear what you already own, and tells you why it picked the bottle it did.

What "shelf-first" means

When you ask Fragrance Engine "what should I wear today?", it does not search the catalog. It searches your shelf. The recommendation engine scores every bottle you own against the current weather, the occasion, what you wore recently, and your past feedback — and then it explains the top pick in plain English.

Try it on a citrus like Bergamot, a powerhouse base note like oakmoss, or the gourmand stalwart Vanilla, and watch how it shifts the recommendation depending on context.

Why this works

Three principles shape the product:

1. Shelf-first — decisions are made over what you actually own.
2. Explainable — every pick comes with a concrete, human-readable reason.
3. Local-first — embeddings happen async; the rules-based engine never keeps you waiting.

If you collect from a single house — say Chypre classics — the engine learns your bias and folds it into the score.

What's next

Over the next few weeks we'll publish deep-dives on layering, designer profiles, and the rules-based scoring engine itself. If you have questions, the Discord link below is the fastest way to reach the team.