[nom]

Data Scientist & Big Data Engineer

Casablanca · | · [e-mail] | [téléphone] · | · [lien] · |

[lien]

Profile

Data Scientist and Big Data Engineer, Master’s graduate in Big Data and Internet of Things, with industrial experience in the automotive sector. Skilled across the full data chain, from data quality and preparation to feature engineering, predictive modelling and the deployment of reproducible pipelines on cloud platforms. Recognised for turning complex business requirements into reliable data products, in close collaboration with domain experts. Seeking a Data Scientist or Data Engineer position within a demanding technical environment.

Education

Master’s Degree in Big Data and Internet of Things · 2024 – 2026

École Nationale Supérieure d’Arts et Métiers (ENSAM) · Casablanca, Morocco

Bachelor’s Degree in Data Infrastructure, Processing and Analytics · 2023 – 2024

École Supérieure de Technologie (EST) · Fquih Ben Salah, Morocco

BTS in Information Systems Development · 2021 – 2023

Centre de BTS · Laâyoune, Morocco

Baccalaureate in Life and Earth Sciences · 2020–2021

Prince Moulay Rachid High School · Es-Semara

Professional Experience

Data Science & Big Data Engineering Intern · April 2026 – Present

Stellantis – Africa Technical Center · Casablanca, Morocco

Subject: Machine learning models for improved car audio NVH performance assessment

• Built the first dataset linking subjective jury ratings to objective Audio NVH measurements, the prerequisite for any predictive model: both sources are produced by different engineering teams and share no common identifier.
• Data audit and recovery: structured 351 audited evaluations into four dependency-linked groups (metadata, expert assessments, derived scores, target ratings) and defined six expert-validated recovery rules; recovered 155 records and raised the usable base from 103 to 258 evaluations (29.3% to 73.5%), avoiding the 71% loss of complete-case deletion.
• Record linkage: designed two matching strategies for the Stellantis and competitor scopes, reproducing expert reason- ing (vehicle model, audio setup, radio system) through exact keys, a four-level progressive search and traceable diagnostic rules; matched 87 of 211 evaluations (41.2%): 47 Stellantis (33.6%) and 40 competitors (56.3%), 34 automatic and 53 validated by domain experts.
• Unit of observation: defined one measurement report (18 frequency-response curves of 119 points) as one sample, preventing label duplication and data leakage, and consolidated an analysis dataset of 46 vehicles at 65 dB on front and rear listening positions.
• Acoustic feature engineering: derived five feature families from the response curves (raw levels, band statistics over six schemes, spectral peaks, psychoacoustic descriptors, BFCC/MFCC cepstral coefficients), compressing 238 raw variables into 26 while preserving perceptual meaning.
• Industrialization: implemented the full chain as a medallion pipeline on Databricks (Bronze, Silver, Gold with Delta Lake and Unity Catalog), producing 17 feature tables through a fully reproducible end-to-end run.
• Technologies: Python, Pandas, NumPy, PySpark, Databricks, Unity Catalog, Spark SQL, Matplotlib, Seaborn.

Data Analyst Intern · August 2024

Ilem Group · Casablanca, Morocco

• Designed Power BI dashboards for HR KPI monitoring and operational reporting.
• Built dimensional models and developed DAX measures for year-over-year analysis, rolling averages and performance tracking.
• Cleaned, transformed and modeled business data using Power Query, Excel and SQL Server.
• Technologies: Power BI, DAX, Power Query, SQL Server, Excel.

Data Scientist Intern · April 2024 – June 2024

Isdy · Casablanca, Morocco

• Developed a predictive maintenance prototype using machine learning models on IoT sensor data.
• Performed data preprocessing, feature engineering, model comparison and hyperparameter tuning.
• Built a Streamlit application for visualizing predictions and supporting user interaction.
• Technologies: Python, Scikit-learn, XGBoost, Random Forest, Pandas, NumPy, Streamlit.

Technical Projects

AG-FIRMA: Agentic AI Application for Smart Agriculture · 2025

ENSAM / Gustave-Eiffel University Collaboration

• Designed a multi-agent LLM architecture for agricultural assistance, including irrigation advice, disease detection and contextual recommendations.
• Integrated a RAG-based knowledge layer to retrieve information from document sources and support explainable re- sponses.
• Developed multimodal and multilingual capabilities using text, image and voice inputs in Arabic, Darija, French and English.
• Technologies: LLMs, RAG, NLP, Computer Vision, Python, APIs, IoT concepts.

Vehicle Detection System using CNN and YOLOv8 · 2025

• Developed a computer vision system for vehicle detection and classification using deep learning.
• Preprocessed more than 120,000 images from multiple datasets, including data balancing and augmentation.
• Trained a custom CNN architecture with convolutional blocks, batch normalization, dropout and regularization.
• Implemented YOLOv8 with transfer learning for real-time multi-vehicle detection.
• Achieved 96.69% accuracy on the classification task.
• Technologies: Python, TensorFlow, Keras, YOLOv8, OpenCV, Streamlit, Google Colab.

Real-Time Data Lakehouse with Medallion Architecture · 2026

• Designed a real-time data lakehouse using Bronze, Silver and Gold layers.
• Implemented streaming ingestion with FastAPI, Kafka and Spark Structured Streaming.
• Used Apache Iceberg and Nessie for table format management and versioning.
• Built an analytics layer using Trino and Apache Superset on S3-compatible MinIO storage.
• Technologies: Apache Spark, Kafka, Apache Iceberg, Nessie, Trino, MinIO, Docker, Superset.

Sentiment Analysis on Financial Text Data · 2024

• Built a sentiment classification pipeline on 4,845 financial sentences from the Financial PhraseBank dataset.
• Implemented text preprocessing, stemming, TF-IDF vectorization and feature extraction.
• Compared several supervised learning models and achieved 74% accuracy.
• Developed an interactive Streamlit interface for real-time sentiment prediction.
• Technologies: Python, NLTK, Scikit-learn, Pandas, TF-IDF, Streamlit.

Technical Skills

Data Science & Machine Learning: Python, Pandas, NumPy, Scikit-learn, XGBoost, Random Forest, AutoML, Feature Engineering.

Deep Learning & AI: TensorFlow, PyTorch, CNNs, YOLOv8, Computer Vision, NLP, LLMs, RAG.

Big Data & Distributed Processing: Databricks, Delta Lake, Unity Catalog, Apache Spark, PySpark, Kafka, Hadoop.

Data Engineering: ETL/ELT pipelines, Data Integration, Data Quality, Record Linkage, Data Modeling, Lakehouse

Architectures, Apache Iceberg, dbt, Airflow.

Databases & Querying: SQL, PostgreSQL, MySQL, SQL Server.

Visualization & Applications: Power BI, DAX, Streamlit, Matplotlib, Seaborn, Sweetviz, Apache Superset.

Cloud & DevOps: AWS Foundations, Microsoft Azure, Docker, Git, Linux.

Certifications

• IBM – Python for Data Science, AI & Development.
• DataCamp – Machine Learning with Tree-Based Models.
• DataCamp – Supervised Learning with Scikit-learn.
• AWS – AWS Cloud Foundations.

Languages

Arabic: Native · | · French: Professional fluency · | · English: Technical and professional working proficiency

Additional Information

Interests: AI research, automotive innovation, data communities, GDG, Enactus, technical podcasts, football refereeing.