Svetlana S.

Designing and developing data pipelines in Microsoft Fabric for end-to-end data ingestion, transformation and loading (ETL). Preparing and structuring data for business intelligence reports and dashboards, ensuring accuracy and consistency across multiple data sources. Working with Spark notebooks to clean, validate and transform large datasets. Collaborating with cross-functional teams to automate data processes and improve the reliability of analytical workflows. Analyzed data and visualized results using histograms and other plots. Tokenized texts to identify key words. Performed data clustering and similarity analysis. Cleaned datasets and applied sampling techniques to select optimal text samples.

Developed web applications using Django. Maintained software for sepsis detection in hospitals. Worked on machine learning and classification problems, including: Calculating probabilities and extracting key features. Developing and optimizing classifiers such as MLP, KNN, XGBoost, SVM (linear and nonlinear problems).

Analyzed data and visualized results using histograms and other plots. Tokenized texts to identify key words. Performed data clustering and similarity analysis. Cleaned datasets and applied sampling techniques to select optimal text samples.