Alexandre Leopoldo

At Fraudio, I build and maintain distributed data pipelines for fraud detection, processing large-scale transactional datasets with Apache Spark, Python, and Scala.

I orchestrate pipelines with Apache Airflow and Databricks jobs to support fraud analysis. My work combines scalable data engineering with practical machine learning systems.

Previously at the Center for Responsible AI at CISUC, I researched Active Learning, human-in-the-loop annotation, and multimodal LLMs for image classification. I also co-authored “Ground Truth Independent Active Learning with Multiple Annotators,” accepted at the HLDM Workshop at ECML-PKDD 2025.