Summary
Victor Suciu is a Data Scientist III with nine years of experience specializing in financial fraud detection and underwriting, currently based in Madison, WI. He has a strong track record of improving ML systems in production—boosting NLP fraud detection F1 by 27% with fine-tuned BERT models and uncovering real fraud rings via graph analysis. Victor combines classical ML (XGBoost, symbolic regression) with deep learning (PyTorch, attention RNNs) and production tooling (Airflow, MLflow, Docker, GCP) to move models from prototype to scheduled services. His background spans academic research in bioimage segmentation and large-class-imbalance credit modeling, giving him a rare blend of domain rigor and practical deployment experience. He’s effective translating technical insights to non-technical stakeholders through clear visualizations and automated pipelines that reduce analyst workload. Outside core fraud work, he’s applied multi-GPU TensorFlow training and graph-based OCR extraction—skills that hint at both systems-level thinking and creative data engineering.
9 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at University of Wisconsin-Madison
B.S., Computer Science & Software Engineering, B.S., Computer Science & Software Engineering at University of Washington Bothell