The ritual begins with one mirror photo.
The AI personal stylist that styles people from the clothes they already own — and earns money the moment it recommends what they genuinely lack.
Karaoglu LTD · Investor briefing · 2026 · Scroll — the deck reads like the app.
The verdict, as the app prints it.
One photo. The on-device engine reads the pieces, scores the look against today's weather and occasion, and names the ONE change that would lift it — on the card, in plain words, never a black box.
The seal is the look's number for today's weather and occasion. Under it: the one change — the single swap that lifts the number most — and why the look works, axis by axis.
The number here is the demo look's, as the capture beside this card prints it; a real one is computed from her closet, never typed in.
of global carbon emissions come from fashion — more than aviation and shipping combined (industry estimates).
the average garment is worn before being discarded (industry estimates) — closets are full of unused value.
the "what do I wear?" decision is made under time pressure, and usually answered by buying more of the same.
Full closet, nothing to wear: fashion e-commerce optimizes for selling more clothes. Nobody optimizes for the wardrobe people already have — the place where the daily decision actually happens.
A closet that knows itself.
Photograph any garment: the background is cut on the device, then the garment is named and classified. The closet becomes data she owns — on her phone first, in the cloud only when she asks.
In-browser background removal (self-hosted models, no cloud cost), silhouette classification trained on 54K labeled garments (MIT-licensed data), the cloud eye (Google's model through our own server) when the device is unsure.
Worn and resting pieces, cost per wear, the colour story, the one gap the calendar is about to expose — computed from what she actually wears, never guessed.
Offline-first; cloud roaming is her opt-in. The mirror photo leaves the phone only for the fitting she asks for and is kept nowhere.
Quality bar: 3473 automated engine tests · 7 end-to-end persona acceptance suites (including a real scan-pipeline test and an offline-resilience test) · zero-backend architecture that deploys to any static host.
The Daily Dose — scored, explained, from her own pieces.
Real styling theory, computed — the same craft a trained stylist applies, and every score explained on the card. Same closet, same day, same answer.
Hue-wheel colour harmony with warm/cool undertone cohesion · dress-code coherence against the occasion · weather · learned taste · proportion · pattern play · freshness. Deterministic, unit-guarded, surfaced in the user-facing "Why?".
Every swipe, wear, save and shuffle nudges a decaying affinity model — including which exact pieces are worn together. The Stylist Brain shows what it learned.
Conversational styling that remembers sizes ("my size is M"), answers sizing and colour questions, and honestly declines what it can't do.
Generative AI fitting (a state-of-the-art image model) paints the outfit onto her own photo. Metered — the one feature that costs us money — and said so.
The full algorithm specification — formulas, constants, update rules — lives in the design document (§4), maintained to due-diligence grade.
Her days in style — the record that makes a circle worth joining.
Style Circles make friends' closets borrowable. When a friend owns the missing piece, the app says borrow — €0, zero new clothes, real impact math (CO₂e, water, €) from saved looks.
When the closet lacks nothing, the shopping section says so and sells nothing. An app that tells you not to spend is the one you believe when it says buy.
Borrow-before-you-buy only works with friends inside — the core feature is the referral mechanic, and every invite is rewarded in-app.
The Stylist Brain shows users every signal the model consumed and what it learned. No black box, no dark patterns.
Partner links are labeled, commissions disclosed ("never at extra cost to you"), cloud-gated features say so instead of simulating.