Project Overview

• Develop a personalized recommendation algorithm that adapts to user behavior and improves relevance over time.
• Focus on implementing and comparing algorithms such as Collaborative Filtering, KNN and SVD for ranking and prediction.

Technical Stack & Responsibilities

• Implement models using Scikit-learn and TensorFlow; perform feature engineering and model selection.
• Process and prepare data with Pandas and NumPy; design efficient data pipelines for training and offline evaluation.
• Containerize components with Docker and prepare deployment workflows for Kubernetes-based infrastructure.
• Produce visualizations and analysis using Matplotlib and Seaborn to communicate model performance and user-facing metrics.

Deliverables & Evaluation

• A working recommendation prototype demonstrating personalized suggestions for sample users.
• Quantitative evaluation (e.g., RMSE, precision@k, recall@k, A/B or offline ranking metrics) and qualitative analysis of recommendations.
• Documentation of experiments, architecture, and instructions to run the system in Docker/Kubernetes.

Candidate Profile & Skills

• Good understanding of recommender systems concepts: collaborative filtering, neighborhood methods (KNN) and matrix factorization (SVD).
• Practical experience with Python, Scikit-learn, TensorFlow, Pandas and NumPy; familiarity with Docker and basic Kubernetes concepts.
• Ability to analyze results and produce clear visualizations and reports using Matplotlib/Seaborn.

How to Apply

• To apply, send your CV and a short motivation letter to issam@code-reason.com.
• In your email, mention relevant projects or coursework on recommender systems and any code samples or GitHub links you can share.