Led a team of seven in the design, development, and deployment of a fraud detection tool (Quantexa) for a leading European bank, leveraging Scala, Spark, Python, PySpark, Hive, AWS, and TypeScript.
Successfully optimized Spark jobs, reducing ETL batch time by 80% and AWS infrastructure costs by 70%.
Developed a data reconciliation and data quality dashboard processing 20+ data sources with over a million records each, integrating multiple platforms like Informatica, Hadoop, and Elasticsearch using Scala, Spark, Python, and Shell Scripting.
Built advanced analytics dashboards using Tableau and Python, modeling financial data to detect fraud risks and new business opportunities.
Spearheaded the development of predictive models using Python and PyTorch, improving fraud detection accuracy by 15%.
Led the development of an anti-money laundering (AML) application for a multinational bank, utilizing Scala, Spark, Hive, and Big Data technologies.
Designed and implemented scoring frameworks for high-risk entities, integrating data from employees, customers, BVD, Panama Papers, and other sources to enhance fraud detection capabilities.