10/04/2026 updated
100 % available
Data Scientist | Machine Learning Engineer
CASABLANCA, Morocco
Worldwide
Engineering Degree in Artificial IntelligenceRAGGCPArtificial IntelligenceData MiningMachine LearningNatural Language ProcessingFeature EngineeringData ScienceTransfer LearningRandom ForestMachine Learning OperationsUnsupervised Learning
Machine Learning & Model Development
Expertise in supervised and unsupervised learning, ensemble methods, feature engineering, model evaluation, and hyperparameter tuning using algorithms such as Logistic Regression, Random Forest, XGBoost, Neural Networks, CNNs, and Transformers.
NLP & Retrieval-Augmented Generation (RAG)
Proficiency in text classification, embeddings using SentenceTransformers, RAG system design with Qdrant vector databases, and LLM integration to reduce hallucination rates and improve semantic retrieval precision.
MLOps & Cloud Deployment
Hands-on experience deploying ML services using FastAPI, Docker, and Kubernetes on GCP, with MLOps best practices including model versioning and experiment tracking via MLflow and W&B.
Recommendation Systems
Building and optimizing content-based and collaborative filtering recommendation systems, incorporating behavioral signals to increase user engagement and reduce cold-start impact.
Computer Vision
Application of transfer learning with VGG-16, VGG-19, and YOLOv5 for image classification and object detection tasks, achieving high classification accuracy on real-world datasets.
Data Engineering
Working with SQL, ETL pipelines, Firestore, and vector databases such as Qdrant, as well as cloud services including GCP, GKE, Cloud Run, and Vertex AI.
Modeling & Analytics
Feature engineering for structured and unstructured data, model evaluation using ROC-AUC, F1-score, RMSE, and NDCG, cross-validation, hyperparameter tuning, class imbalance handling, and statistical hypothesis testing.
Time-Series & Behavioral Data Modeling
Modeling time-series and behavioral data, integrating real-time user interaction signals to dynamically update ranking models.
Expertise in supervised and unsupervised learning, ensemble methods, feature engineering, model evaluation, and hyperparameter tuning using algorithms such as Logistic Regression, Random Forest, XGBoost, Neural Networks, CNNs, and Transformers.
NLP & Retrieval-Augmented Generation (RAG)
Proficiency in text classification, embeddings using SentenceTransformers, RAG system design with Qdrant vector databases, and LLM integration to reduce hallucination rates and improve semantic retrieval precision.
MLOps & Cloud Deployment
Hands-on experience deploying ML services using FastAPI, Docker, and Kubernetes on GCP, with MLOps best practices including model versioning and experiment tracking via MLflow and W&B.
Recommendation Systems
Building and optimizing content-based and collaborative filtering recommendation systems, incorporating behavioral signals to increase user engagement and reduce cold-start impact.
Computer Vision
Application of transfer learning with VGG-16, VGG-19, and YOLOv5 for image classification and object detection tasks, achieving high classification accuracy on real-world datasets.
Data Engineering
Working with SQL, ETL pipelines, Firestore, and vector databases such as Qdrant, as well as cloud services including GCP, GKE, Cloud Run, and Vertex AI.
Modeling & Analytics
Feature engineering for structured and unstructured data, model evaluation using ROC-AUC, F1-score, RMSE, and NDCG, cross-validation, hyperparameter tuning, class imbalance handling, and statistical hypothesis testing.
Time-Series & Behavioral Data Modeling
Modeling time-series and behavioral data, integrating real-time user interaction signals to dynamically update ranking models.
Experience
Professional experience0 - 5 years
Freelancer since2026
Languages
EnglishFluentFrenchFluent
Project history
Designed and deployed a RAG system using SentenceTransformers and Qdrant. Built and optimized recommendation systems. Engineered ML pipelines for grammar rule classification and flashcard ranking. Deployed ML services using FastAPI and Docker on GCP with Kubernetes. Applied MLOps best practices including model versioning and experiment tracking with MLflow and W&B.
Built intelligent features for an e-learning mobile application (Leyra). Developed a hybrid recommendation system combining content-based filtering and collaborative filtering. Applied feature engineering on user interaction data. Integrated NLP capabilities including automatic summarization and conversational chatbot.
Built a fraud detection system using supervised classification models including Random Forest and XGBoost. Applied class imbalance handling techniques such as SMOTE and class weighting. Reduced false positives by 10% and improved detection latency by 25%. Evaluated models using ROC-AUC, Precision-Recall curves, and confusion matrix analysis.