At KTH Lab, National University of Sciences and Technology (NUST), I engineered a forensic pipeline that extracts 20+ features per session from filesystem, registry, network, and memory evidence. I also built a Python toolchain of 25+ scripts to automate evidence capture, extraction, validation, and model training.
I built a machine learning pipeline that classifies six browser activity types across Normal, Incognito, and Tor environments. I validated datasets through automated checks and manual row-level audits, and correlated Windows Security Event Logs with filesystem and network evidence to corroborate credential-entry events.
At Intelligent Learning Machines, Inc., I supported cybersecurity risk-assessment work aligned with ISO/SAE 21434 for automotive systems. My other work includes solving Hack The Box machines and building a secure messaging application and smart-contract simulation in JavaScript.

