Lathika Ekambaram

I developed an ML-based intrusion detection system using NSL-KDD and CIC-IDS2017 network traffic datasets. It classifies activity as normal or malicious with Random Forest and XGBoost, achieving over 95% detection accuracy.

For an automated Cyber Threat Intelligence pipeline, I used Python to collect, normalize, categorize, and enrich indicators of compromise. I mapped indicators and observed behaviors to MITRE ATT&CK and generated reports to support analyst investigations.

I built a SIEM-based lab to collect and analyze Linux and network security logs. I created detection rules and queries, correlated host and network events, and used Python automation to highlight suspicious activity.

At Centre for Digital Health and Precision Medicine (CDHPM), I worked as an AI Intern, developing a machine-learning model for prediction and data analysis. I also have a strong interest in Cyber Threat Intelligence, security automation, detection engineering, incident investigation, and AI-assisted cybersecurity workflows.