Скачать или смотреть SearchOnQdrant by EarlyThanThis

SearchOnQdrant: Building a Semantic Search Engine with Vector Similarity
By team - EarlyOnThis

I built a complete semantic search engine using Qdrant vector database! This project demonstrates how to create intelligent search that understands meaning, not just keywords.
What it does:
Converts text to 384-dimensional vectors using Sentence Transformers
Performs real-time vector similarity search with Qdrant
Includes geotgging and location-aware filtering
Delivers sub-second search responses with rich metadata

🛠️ Tech Stack is your good ol
Backend: FastAPI + Qdrant + Sentence Transformers
Frontend: Next.js 14 + TypeScript + Tailwind CSS
Database: Qdrant vector database with Docker
Key Features:
Semantic search with cosine similarity
Geographic intelligence here we wnet ahead and used the Haversine distance calculation
Real-time performance (is around 250ms average)
logging mostly for testing and monitoring

How the Results look like :
20 sample documents with rich metadata
Vector similarity scores (0.0 to 1.0)
Location data with GPS coordinates
Categories, tags, and timestamps

GitHub:https://github.com/Cappybara12/QdrantHack