Data Scientist | Statistics, Machine Learning & Risk
📍 Reims, France (MSc SEP, University of Reims)
🎯 Turning rigorous statistics into models that run in production
I hold a statistician-analyst degree from ENSAE Pierre Ndiaye (Dakar) and I am now an MSc student in Statistics for Evaluation and Forecasting (SEP) at the University of Reims, France.
My work sits at the intersection of rigorous statistical methods and applied machine learning: anomaly detection, multi-label classification, large-scale data processing, and shipping models to production instead of leaving them in notebooks. Before that, I spent three years designing surveys, running data collection and supervising field teams, which taught me that a model is only as good as the data pipeline behind it.
Short-term goal : an apprenticeship in data science within the banking sector for the 2027 intake, on risk modelling, fraud detection or credit scoring.
Long-term goal : contribute to AI safety research, making powerful AI systems more reliable, interpretable and aligned with human values.
| Project | Description | Stack | Topics |
|---|---|---|---|
| malware_classification | Multi-format threat detection (Windows PE, PDF, URL, Office/VBA), up to 99.82% accuracy, four independent serverless microservices. | Python · LightGBM · XGBoost · Optuna · AWS Lambda | Anomaly Detection · ML Engineering · Cybersecurity |
| toxic-comment-classification | Multi-label toxicity detection with RoBERTa and XLM-RoBERTa, deployed on AWS with a live dashboard. | Python · HuggingFace · AWS | NLP · Transformers · Content Moderation |
| spark-energy-weather-analysis | 590K observations across 32 European countries: how weather extremes drive electricity consumption. | PySpark · SciPy · Jupyter | Big Data · Data Engineering |
| yolov8-detection-studio | Interactive object detection studio for images, webcam and video, with a live demo on Hugging Face. | PyTorch · Ultralytics · Gradio | Computer Vision · Deep Learning |
- 🎓 Studying : statistical inference, stochastic processes, neural networks, SAS and SQL (MSc SEP, M1)
- 🏦 Preparing : an apprenticeship in banking data science for September 2027
- 📖 Reading : Constitutional AI (Anthropic, 2022) and Attention Is All You Need
- 🌍 Interested in : NLP for African languages (Wolof, Bambara) with multilingual transformers
Languages : Python · R · SQL · SAS · PySpark
ML / NLP : Scikit-learn · XGBoost · LightGBM · Optuna · HuggingFace Transformers · PyTorch
MLOps : AWS (Lambda, S3, SageMaker) · Docker · Git
Data : Pandas · NumPy · Plotly · Dash · Stata · ODK
2026 ──── MSc SEP (M1), University of Reims · Portfolio consolidation
2027 ──── M2 SEP as an apprentice · Data science in banking
2028 ──── First data scientist role · Risk, fraud or scoring
2032 ──── Senior data scientist, or research through an industrial PhD
2035 ──── Research engineering in AI safety, the long-range target
"The goal of AI safety is not to slow AI down: it is to make sure that when we get there, it was worth arriving."