Freelancer profile translated to English.
Back to original languageAbout Baptiste Marc
**I develop solutions that optimize the use of your data on an efficient pipeline to strengthen your competitiveness**.
**Experience**:
Data Science
Data Processing
— Python (poetry) / R
— ETL (Airflow, DBT, Fivetran)
— Data Analysis / EDA
— Stream (Beam, Dataflow, Kafka)
— Hadoop / Snowflake
— PySpark
Data Collection
— Data Engineering / UML
— Hadoop / Spark
— Web Scraping (Beautifulsoup, Selenium, Scrapy)
— LLM RAG / ChatGPT
LLM / ChatGPT API
— Prompt Engineering (CoT & Tool Calling)
— RAG (LangChain, LlamaIndex)
— Vector DB (FAISS, Chroma)
— Agents & Tool Use
— Guardrails, HPO & Monitoring
Prediction & Classification
— Modeling
— Statistical, AI & Machine Learning Models (50+ models)
— Deep Learning (Tensorflow)
— PyTorch, ScikitLearn
— Computer Vision
— Time Series
— NLP
Monitoring / Alerting
— Dashboards (PowerBI, Tableau, Looker, Shiny, Dash)
— Data Visualization
— ELK Stack
Product Development
Prototyping
— UX/UI & Research (Figma, Adobe XD)
Development
— Front (React, JS, HTML, SCSS)
— Back / APIs (Flask, FastAPI)
— Advanced SQL / NoSQL
— Automation (N8n, Make, Zapier)
Production
— CI/CD: GitHub, Jenkins, Serverless Framework
— Cloud: GCP / AWS (Lambda, API Gateway, S3, RDS)
— TensorFlow, PyTorch, MLflow, Scikit-learn
— Databricks, Dataiku,
— Docker, Kubernetes
— Cybersecurity (OWASP, ISO27001, EBIOS)
- 📊Lead Senior Data Scientist(9 years)
- 🔬AI Researcher / Mathematics
- 🏛️Professor of Data Science(Catholic University of Lille)
- 🦄AI SaaS Founderweeki.io (5 years at Euratechnologies)
Education:
- PhD Student in Machine Learning
- Trained in the USA in NYC & Stanford 🇺🇸
English
Native or bilingual
French
Native or bilingual
Greek
Native or bilingual
Can work on-site
Paris (up to 50km), Lille (up to 50km), Ferney-Voltaire (up to 50km), Thonon-les-Bains (up to 50km)
Experience
- DECATHLON | TECHNOLOGYSenior Data Scientist for Supply Chain ForecastRETAIL (LARGE RETAILERS)January 2024 - January 2025 (1 year)Paris, FranceLead Senior Data Scientist – Forecast & Supply ChainMissionDemand forecasting, inventory optimization, and replenishment.Models adapted to seasonal variations.Team: 1 PM, 4 DS, 2 DA, 1 MLContextRetail, complex supply chain, e-commerce, omnichannelRobust modeling for demand variability, consumer behavior, supply constraintsObjectivesMulti-scale forecastingProbabilistic forecasts (distributions, confidence intervals)Incremental learning (cutoffs)Deployment automationInventory optimization, reduction of stockoutsTechnologies & InfrastructureLanguages: Python (Pandas, Numpy, PySpark, Scikit-learn, TensorFlow)Frameworks: FastAPI, Streamlit, PoetryCloud & Infra: Databricks, AWS S3, GitStorage & Pipeline: Colibra, Delta files, Parquet, Databricks JOBS templateOrchestration: AirflowCI/CD & DevOps: GitHub Actions, DockerML Management: MLflow, DatabricksMonitoring & Viz: TableauData & PipelinesCentralization on Colibra / S3 / ParquetData quality validated before ingestionSources: sales history, economic indicators, external signalsTransformations & AnalysesClustering, seasonality, PCA, statistics (Anova, Chi-squared, T-test, F-Fisher)Feature Engineering: lags, macroeconomics, KNN, clusteringModeling & PredictionFeature Selection: SelectKBest, Boruta, RFE, SFS, Random Forest ImportanceML Forecast: LightGBM, XGBoost, RandomForest, CatBoost, AdaBoostTime Series: SARIMA, TFT, RNN, STL, ARIMA, Fourier, Seasonal PolynomialOptimization: HPO (Optuna, HyperOpt)Cost Functions: RMSE, WAPE, MAE, Tweedie, Quantile LossFeature ImportanceLocal: SHAP, LIMEGlobal: Beta Coefficients, Friedman H, PermutationDeployment & OperationsAPI via FastAPI, UI via Streamlit, hosted on Databricks APPAirflow automation, error/drift monitoringDrift alertsReporting & VisualizationTableau DashboardsComparative BacktestingScenarios
- UPFUNDSenior Lead Data Scientist for Real EstateREAL ESTATEJanuary 2024 - January 2025 (1 year)Paris, FranceMachine Learning (ML) →- Prediction of real estate indicators (commercial, apartments, houses) using Machine Learning models, geospatial analysis, and time series forecasting.Data Engineering (DE) & Data Analysis (DA) →- Creation of data pipelines, preprocessing, exploratory data analysis (EDA)Research & Development (R&D) →- Identification and definition of research problems to guide projects in a structured and scientific manner. Definition of working hypotheses, with production of summaries of the models used.Knowledge Management (KM) →- Creation of a state-of-the-art on spatial statistics, time series, forecasting, and Machine Learning applied to real estate.- Centralization, structuring, and management of scientific knowledge to leverage expertise and facilitate its reuse.
- UNIVERSITE CATHOLIQUE DE LILLEProfessor in Datascience / ML / Probability & StatisticsBIOTECHJanuary 2023 - January 2025 (2 years)Lille, FranceCourse Program – Visiting Professor in MathematicsFoundations• 0.1: Elements of Calculus and Tools• 0.2: Epistemology and Theory of KnowledgePart 1 – Systems Theory• 1.1: Dynamical Systems• 1.2: Complex Adaptive SystemsPart 2 – Stochastic Dynamics and Probabilities• 2.1: Measure Theory• 2.2: Probability Theory• 2.3: Common Probability Distributions• 2.4: Asymptotic Statistics• 2.5: Stochastic Processes and Time Series• 2.6: Information GeometryPart 3 – Data Observation• 3.1: Descriptive Statistics• 3.2: Exploratory Data AnalysisPart 4 – Inference and Estimation Theory• 4.1: Parameter Estimation• 4.2: Experimental Design, Sampling, and Hypothesis Testing• 4.4: Decision Trees and Model Selection• 4.5: Bayesian InferencePart 5 – Examples of Linear and Regression Models• 5.1: Simple Linear Regression• 5.2: Multiple Linear Regression• 5.3: Other Regression MethodsPart 6 – Other Examples of Classical Models• 6.1: Common Univariate Tests• 6.2: Common Multivariate Tests• 6.3: Non-parametric StatisticsPart 7 – Examples of Non-linear Models• 7.1: Probabilistic Graphical Models• 7.2: Percolation Theory• 7.3: Spatial Statistics• 7.4: Extreme Value Theory• 7.5: Agent-Based Modeling• 7.6: Network DynamicsTechnologies and tools used:MATLAB, R, Python, LaTeX, Jupyter Notebooks, SPSS, SAS, Excel, NumPy, SciPy, Pandas, Matplotlib, TensorFlow, PyTorch, Tableau, Power BI, SQL, GitHub.
Recommendations
FU
+9
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
- MASTER in MATHEMATICS and COMPUTER SCIENCE applied to COGNITIVE SCIENCE for BUSINESSLille University2019— Data Science with Python: Machine Learning — Probability — Statistical Linear Models & Regression — English M1 — Web — Computing for Neurocognitive Science — Digital development for Neuropsychology — Philosophy of Mind — Ergonomy & Product design — R programming — Non Parametric Statistics — SAS for datascience — E-marketing — Technology for Psychological Research M2 — Ethics & deontology — Functionnal Neuroscience — Emotionnal Process & Affective neuroscience — Neurocognition — Artificial Neural Networks — Programming for Experimental research — UX design / Product and Experience optimization
- NEW YORK CITY DATA SCIENCE ACADEMYNew York Datascience Academy (NYCDSA)2019— Deep Learning — Statistical models — Hadoop — Spark — AWS — Datavizualizatiuon — Linux system — Advanced SQL — NoSQL — Web Scraping — Time Series Analysis — Reinforcement Learning — Computer Vision — Generalized Linear Models — Tree Methods — Support Vector Machines — Natural Language Processing — Code Optimization — Advanced Phyton — Advanced R