Description

💡Qu'est-ce que je fais?

Je transforme vos données brutes en modèles intelligents qui créent de la valeur réelle pour votre business.

👩🏻‍💻 Qui suis-je?

Data Scientist & Machine Learning Engineer (4+ ans d'expérience), diplômée de l' ENS Ulm et de l'École Centrale de Lyon.

Je conçois des solutions ML robustes et sur-mesure, de la recherche jusqu'au déploiement en production.

📊 Mon approche?

Comprendre vos enjeux métier avant tout, puis construire le modèle le plus adapté pour y répondre avec précision.

📋 Ce que je fais concrètement?

Machine Learning & Modélisation

Modèles prédictifs sur-mesure : scoring, segmentation, classification, régression, séries temporelles, détection d’anomalies, systèmes de recommandation.
Construction rigoureuse des données d’entraînement : stratégies de sampling (stratifié, pondéré), gestion du class imbalance, détection et prévention du data leakage.
Feature engineering avancé : encodage de variables catégorielles, feature crossing, sélection par feature importance et généralisation.
Évaluation et sélection de modèles : baselines, ensembles, validation croisée, hyperparameter tuning.

NLP & IA Générative

Analyse de texte, classification de documents, extraction d'entités.
Fine-tuning de modèles de langage (BERT, Transformers, Hugging Face).

💼 Mon expérience?

AdTech (Adikteev), Assurance (SCOR), Cybersécurité (Institut Polytechnique de Paris), Supply Chain (Danone) — des contextes variés qui me permettent de m'adapter vite à vos problématiques métier.

🛠️ Stack?

Python, PyTorch, TensorFlow, Scikit-Learn, Hugging Face, SQL, Spark...

📍 Basée à Paris | Remote ou hybride | Je réponds vite.

Industry field of expertise

Languages

English
Native or bilingual
French
Native or bilingual
Arabic
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 20km)

Adikteev
Machine Learning Engineer
March 2022 - Today (4 years and 4 months)
Paris, France
○ Collaborated cross-functionally with engineering, product, and data teams to translate business constraints into scalable ML solutions deployed in real-time environments.
○ Led the end-to-end development and production deployment of a Bayesian logistic regression model predicting conversion probability in real-time bidding (RTB) auctions.
○ Architected and optimized the end-to-end ML pipeline (data processing, feature engineering, training, validation,
inference), reducing compute cost by 20% and significantly improving latency and system efficiency.
○ Integrated high-impact predictive signals (bidding counters, churn indicators) and engineered preprocessing
strategies to improve production stability.
Machine learning Data science Python
SCOR,
Data Scientist
BANKING AND INSURANCE
May 2021 - November 2021 (6 months)
Paris, France
○ Designed statistical forecasting models to estimate mortality trends and crisis impacts across global demographic
segments using ARIMA-based approaches.
○ Processed and standardized large-scale heterogeneous datasets from international life-table sources spanning age,
country, and temporal dimensions (1,400+ configurations).
○ Presented mortality modeling findings at Longevity 16 (Bayes Business School, Pensions Institute) to actuarial and industry stakeholders as a parallel session speaker.
Data science Statistiques Time Series
Institut Polytechnique de Paris,
Data Scientist intern
ENERGY AND UTILITIES
April 2020 - October 2020 (6 months)
Palaiseau, France
○ Conducted large-scale exploratory analysis on multi-feature datasets (50+ variables), applying dimensionality
reduction (PCA) to identify high-value predictive signals.
○ Engineered features and data pipelines to support robust anomaly detection models.
○ Implemented and evaluated multiple algorithms (K-Means, Isolation Forest, statistical thresholds), improving
detection performance and increasing F1-score by up to 4 percentile points.
○ Delivered interpretable model insights to support decision-making in applied research environments.
Deep Learning LSTM Cybersécurité