Alexandra Siakavella

At EY, I used Python and SQL to extract, transform, validate, and investigate datasets for analytical use. I worked with Azure Databricks and Apache Spark/PySpark to prepare structured datasets for downstream analysis.

I developed and maintained automated data pipelines using Azure Data Factory, incorporating validation and monitoring into data-processing workflows. I also used data-quality checks and systematic troubleshooting to investigate processing errors and support reliable analytical outputs.

In machine learning projects, I applied Random Forest and XGBoost to investigate data patterns and develop predictive models. I analysed features and model outputs to assess predictive results.

On an API data ingestion project, I developed a Python workflow to prepare JSON data for querying and analysis, using Pandas and DuckDB for processing and dbt for reusable transformations and data tests. I’m interested in evaluating AI-generated quantitative solutions, identifying statistical and methodological errors, and improving AI model reliability.