Day 0: Setup and First Steps on Qdrant - Vector Search Engine

Qdrant Setup

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Day 0

Qdrant Setup

Spin up production-grade vector search in minutes. Qdrant Cloud gives you a managed endpoint with TLS, automatic backups, high-availability options, and a clean API.

Create your cluster

Sign up at cloud.qdrant.io with email, Google, or GitHub.
Open Clusters → Create a Free Cluster. The Free Tier is enough for this course.

Pick a region close to your users or app.
When the cluster is ready, copy the API key and store it securely. You can make new keys later from API Keys on the cluster page.

Implementing a Basic Vector Search

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Day 0

Implementing a Basic Vector Search

Follow along as we build your first collection, insert vectors, and run similarity searches. This guided tutorial walks you through each step.

Step 1: Install the Qdrant Client

To interact with Qdrant, we need the Python client. This enables us to communicate with the Qdrant service, manage collections, and perform vector searches.

!pip install qdrant-client

Step 2: Import Required Libraries

Import the necessary modules from the qdrant-client package. The QdrantClient class establishes connection to Qdrant, while the models module provides configurations for Distance, VectorParams, and PointStruct.

Project: Building Your First Vector Search System

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Day 0

Project: Building Your First Vector Search System

Time to apply what you’ve learned. You’ll create a complete, working vector search system from scratch.

Your Mission

Build a functional vector search system that demonstrates the core concepts: collections, points, similarity search, and filtering. You’ll design simple 4-dimensional vectors that represent different concepts or items.

Estimated Time: 30 minutes

What You’ll Build

A working search system with:

One collection with 4-dimensional vectors and Cosine distance
5–10 points with hand-crafted vectors and meaningful payloads
Basic similarity search to find nearest neighbors
Filtered search combining similarity with payload conditions

Setup

Prerequisites

Qdrant Cloud cluster (URL + API key)
Python 3.9+ (or Colab)
Required packages: qdrant-client.

Models

None. We will create vectors by hand.

Dataset

None. We will create our own data points.

Before creating data, decide what each of the four dimensions in your vectors will represent. This is the creative part of vector search!