Teddy Razafindratsima

Capstone Project – Physical Access Control Anomaly Detection System · Developed a machine learning–based anomaly detection system to identify abnormal physical access control events · Used Python for data preprocessing, feature engineering, and predictive modeling · Stored and analyzed anomalous access events in Neo4j, enabling graph-based relationship analysis · Designed ETL pipelines to integrate access control data into analytics workflows · Built interactive Tableau dashboards to visualize anomalies and trends for business and security stakeholders

· Evaluated and compared multiple AI models across diverse use cases to identify the most effective solutions. · Conducted in-depth analysis to determine optimal model performance for specific applications

· Automated the delivery of Credly badges for students who passed the RMF certification. · Designed and implemented a Zapier workflow that reduced manual work and improved efficiency. · Gained experience in automation tools, workflow integration, and credentialing platforms.