Welcome to Gifty Engineering Docs 🎁

This portal contains all technical documentation for the Gifty project — a smart gift recommendation platform. We aggregate product data from popular platforms, analyze it using AI, and provide a user-friendly interface for discovery.

Section	Description	Link
Onboarding	Step-by-step plan for new members: access, chats, first tasks.	Open →
Local Setup	Boot up the entire system (Docker, API, Bot) in 5 minutes.	Open →
Architecture	High-level overview of how data turns into recommendations.	Open →
Data & DB	Data domains, PostgreSQL schemas, and migration rules.	Open →

🏗 Core Systems

🧠 Recommendation Engine

Our SOTA engine based on Qwen2-VL and Vector Search. We don't just search for products; we analyze them across 10 psychological axes (GUTG). 👉 Read more about RecSys →

🕷 Parsing System

Scalable data collection system using Scrapy and RabbitMQ. Features automatic category mapping and AI validation. 👉 How parsers work →

🤖 Telegram Admin Bot

Mission control center. Health monitoring, manual parsing control, and access management. 👉 Bot guide →

🧪 Quality & Monitoring

We believe in TDD and Observability:

Testing: Every feature is covered by pytest. Testing Guide →
Analytics: We track business success via PostHog. Why do we need analytics? →
CI/CD: Automatic deployment on every push. Deployment process →

Any questions?

Reach out in Slack #gifty-dev or message your mentor. We are always happy to help!

Contributing

If you find an error in the documentation or want to add a new section, feel free to create a Pull Request!