Skip to main content
JustSoftLabJustSoftLab
JustSoftLabJustSoftLab
AI Assistant
All case studies
E-commerceAI & GenAIData EngineeringPlatform Engineering

AI-powered personalization engine for e-commerce

Client: European E-commerce RetailerDuration: 5 monthsTeam: 4 engineers

+23%

Conversion rate increase

+18%

Average order value

<100ms

Recommendation latency

2M+

Monthly visitors served

The challenge

A mid-market e-commerce retailer with 2M+ monthly visitors was showing the same product recommendations to everyone. Conversion rates were flat despite growing traffic.

Our solution

We built a real-time personalization engine that combines collaborative filtering, content-based recommendations, and contextual bandits. The system serves personalized product recommendations, search re-ranking, and dynamic homepage content — all updating in real-time based on user behavior.

Key highlights

1

Real-time signals

User behavior updates recommendations instantly

2

Multi-model ensemble

Collaborative + content-based + contextual bandits

3

Search re-ranking

Personalized search results per user profile

4

Dynamic homepage

Content adapts to each visitor in real time

Tech stack

PythonTensorFlowRedisKafkaBigQueryNext.jsGCP
The ROI was clear within the first month. Every visitor now gets a different, relevant experience.

Head of Product

European E-commerce Retailer

Need similar results?

Let's discuss your project.

Get in touch