Schékina MAYENE

– Optimization of the price recommendation pipeline to meet the specific needs of stakeholders (Pricing Owners). Automated data integrity checks, reducing processing errors and accelerating update deployment by 70%. Code versioning and integration managed through Git. – Developed a machine learning model to classify quote line statuses (in range, below, above) with feature engineering in Python, improving price range prediction accuracy. Models used: Random Forest, XGBoost, LGBM (82% accuracy, 74% recall on target class). – Data extracted using SQL from the Snowflake platform. – Built a Power BI dashboard to visualize business solution KPIs (e.g., usage frequency), saving 10 hours per week of manual reporting.

– Built datasets for analysis on the Dataiku platform, reducing data preparation time by 90% – Studied the correlation between data quality KPIs and business performance by applying statistical methods (correlation coefficient, PCA) – Developed a classification model for commercial offers using Logistic Regression in R. Conducted statistical evaluation of the model using hypothesis testing for global coefficient significance.