التطوير، والتحويل الصناعي، وعمليات تعلم الآلة (MLOps)

Data preparation and engineering: Explore, clean, and transform historical data stored in the Lakehouse using PySpark notebooks to design input variables (feature engineering).
Modeling and training: Develop, train, and evaluate several machine learning algorithms (predictive or classification models) adapted to the chosen business problem. Monitoring and
MLOps: Track training cycles, record performance metrics, and package the best model using the native integration of MLflow within the Microsoft Fabric environment.
Industrialization and reporting: Automate scoring pipelines (predictions) via Data Factory to feed the Gold tier of the architecture, and design a Power BI report to visually present predictions and performance indicators to user teams.

Technical Environment:
Data Platform: Microsoft Fabric (Synapse Data Science, Synapse Data Engineering, Lakehouse).
Languages ​​& Data Science: Python, PySpark, specialized libraries (Scikit-Learn, XGBoost, Pandas).
Lifecycle Management (MLOps): MLflow integrated into Fabric.
Orchestration & Reporting: Data Factory (Pipelines), Power BI.

Desired Candidate Profile

B+5 / Engineering school specializing in DATA