Company Performance Metrics
- Sahib Alizada: Founder & CEO
Geyin.az: The AI-First Fashion Marketplace for Azerbaijan & Turkic World
Geyin.az is an AI-native, multi-vendor fashion marketplace built to digitize the underserved fashion retail market across Azerbaijan, Turkey, Russia, and the broader Turkic world. The platform operates in 4 languages (AZ/TR/RU/EN) and serves both consumers
seeking personalized fashion discovery and local vendors who need zero-friction digital storefronts.
What Makes Us Different — The AI Agent Army: We operate 11 autonomous AI agents running 24/7 on a dedicated server, producing fashion intelligence and multilingual content with zero hallucination. Our "Verified Intelligence Pipeline" ensures every piece of published content is scraped from primary sources, cross-verified by multiple agents, and quality-scored by a commander agent before publication. The entire system runs for $55/month.
Core AI Features: - Smart Upload: Vendors photograph a product, and Google Gemini 2.5 Flash automatically generates names, descriptions, categories, colors, sizes, and pricing suggestions in 4 languages — reducing product listing time from 15 minutes to 30 seconds. - Visual Search: Customers upload any fashion photo and find visually similar products using multimodal embeddings (Vertex AI, 1408-dimensional vector space). - AI Content Engine: Daily blog posts, trend reports, brand analyses, and competitor intelligence — all AI-generated with verified sources and zero fabricated facts.
Technical Stack: - Backend: Django 4.2 on Google Cloud Run (auto-scaling, serverless) - AI: Google Gemini 2.5 Flash + Vertex AI multimodal embeddings - Search: Elasticsearch 8.x with dense vector fields for semantic + visual search - Database: PostgreSQL on Cloud SQL - Async: Google Cloud Tasks + Cloud Scheduler (11 scheduled jobs) - Agent Server: 11 Node.js agents on dedicated Ubuntu server with SOUL+Memory system and real-time event bus for agent-to-agent communication
Traction: - 17+ AI-generated blog posts live (4 languages each) - 34 fashion brands tracked autonomously - 10 competitors monitored daily - Full E-E-A-T compliance for AI search engine citation (ChatGPT, Perplexity, Google AI Overviews)
Vision: We are building the fashion intelligence layer for emerging markets — where AI doesn't just list products but understands fashion trends, predicts demand, and creates trustworthy content at scale. Next milestones include custom ML model fine-tuning on verified fashion data and a Dialogflow CX-powered AI fashion assistant in 4 languages.
Founded: November 2024 | Headquarters: Baku, Azerbaijan Founder: Sahib Alizada (linkedin.com/in/sahib-alizada)