Freelancer profile translated to English.
Back to original languageAbout Yann
I help companies reduce their financial losses by detecting anomalies and anticipating risks, with models actually used in production.
Most models are never exploited.
My role: **transform your data into concrete decision-making tools used daily**.
🎯 **I work on**:
- Anomaly detection (fraud, drifts, incidents)
- Risk scoring and modeling
- Forecasting and anticipation (failures, behaviors, trends)
đź’ˇ **Recent examples**:
- Predictive maintenance → reduction of machine downtime
- Scoring → identification of risk profiles
- Monitoring → detection of drifts in production
⚙️ **Approach**:
From analysis to production deployment (**MLOps**), with a constant focus on **business impact and model adoption**.
đź’ˇ **Methodology**: pragmatic collaboration, simple rituals, and concrete monitoring of business impact.
French
Native or bilingual
English
Fluent
Can work on-site
Dijon (up to 50km), Rochefort (up to 50km), La Rochelle (up to 50km), Les Sables d'Olonne (up to 50km)
Experience
- ELITE BEES
On Malt
Data ScientistAUTOMOBILEJanuary 2025 - Today (1 year and 5 months)Recurring data missions on targeted marketing operations from large client bases.My contributions:*Extraction**, **cleaning**, and **standardizationof client databases using Python/Pandas, adhering to complex selection criteria.- Automation of data manipulation, filtering, and export tasks under quality and deadline constraints.
- Delivery of results in ready-to-use Excel formats, with clear documentation for the client.
- Qualitative objectives achieved at 100% (criteria, deduplication) and quantitative objective exceeded by 15%.
- Result: 3 out of 4 clients signed a purchase order following the initial collaborations.
- DataTalks.ClubData Scientist – Data Engineering & MLOpsEDUCATION AND E-LEARNINGSeptember 2024 - August 2025 (11 months)Completed 5 advanced projects within the DataTalks.Club Zoomcamps, focusing on ML model industrialization and data engineering:*MLOps– Full deployment of an ML model with an automated pipeline (Prefect, Docker), experiment tracking (MLflow), and monitoring (Evidently).*Finance – Portfolio Optimization– Cloud pipeline (dbt, BigQuery) for collecting and analyzing financial data, with a decision-making dashboard.*Finance – FinMLOps– Industrialization of a trading model: automated ingestion (yfinance), experiment tracking (MLflow), and CI/CD for reproducible deployment.*Industry– Predictive model for industrial equipment failure based on IoT data, deployed via API.*Finance/Risk– Modeling of corporate financial distress using historical data, with tuning and interpretability of results.
- IndépendantValue Hunter – Investor Data ScientistPRIVATE EQUITYJuly 1994 - Today (31 years and 11 months)Dijon, FranceDevelopment of investmentalgorithmsto grow tax-advantaged portfolios (life insurance, PEA, PEA-PME).My contributions:*Data collectionon ETFs and mutual funds
- Dynamic allocation based on risk profile and asset behavior
- Weekly arbitrage, purging, and rebalancing
- Performance over 8 years: +271% (life insurance), +243% (PEA), +42% (PEA-PME)
Reviews
Recommendations
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Machine Learning ZoomcampDataTalks.Club2025La première partie du cours couvre les algorithmes d'apprentissage automatique implémentés en Python, notamment la régression linéaire, la classification, les arbres de décision, l'apprentissage d'ensemble et les réseaux neuronaux. La deuxième partie se concentre sur le déploiement de modèles à l’aide de frameworks tels que Flask, TensorFlow et Kubernetes, permettant une application pratique de l’apprentissage automatique dans des scénarios réels.
- Data ScientistOpenClassrooms2023Collecter, préparer, explorer et analyser les données avec Python Élaborer des modèles prédictifs et les déployer sur le cloud Communiquer les résultats grâce à des visualisations
Certifications
- Machine Learning ZoomcampDataTalks.Club2025
- Data ScientistOpenClassrooms / CentraleSupélec2023