Mehryar Majd, M.Sc.
PhD researcher · Research assistant
About
Mehryar Majd is a PhD student and research assistant in the Security Engineering group at the Hasso Plattner Institute. The research covers three areas:
- Web-log anomaly detection in the SOC
- SOC threat detection and HTTP log analysis across multiple attack vectors via Web Dispatcher gateways
- Spark-based alerting systems using hybrid detection approaches for large-scale security monitoring
- NLP-driven feature engineering and anomaly detection for identifying XSS, SQL injection, and malicious scanner activities
- Statistical and ML-based anomaly detection on time-series security logs and operational metrics
- Alert quality enhancement through noise reduction, event correlation, and ETL pipeline optimization
- VM workload characterization and prediction in cloud data centers
- AI/ML-Ops approaches for provisioning and resource optimization in cloud data centers and hyperscaler environments through adaptive workload regression and continuous model re-training
- Data sanitization and interval uncertainty modeling for VM workload utilization analysis
- Dynamic right-sizing recommendations using workload forecasting, Compute Instance Behavioral Analysis (IBA), and workload profiling
- LLM security and secure AI-driven application frameworks for cybersecurity
Focus
- Web-log anomaly detection in security operations centres
- Cloud workload prediction
- LLM security
Tags
Publications
When Prompts Become Payloads: A Framework for Mitigating SQL Injection Attacks in Large Language Model-Driven Applications
Proceedings of the 18th International Conference on Agents and Artificial Intelligence (ICAART'26), March 5-7, 2026, Marbella, Spain (AR=129/716)
Large Language Models in Cybersecurity: State-of-the-Art
Proceedings of the 11th International Conference on Information Systems Security and Privacy (ICISSP'25), February 20-22, 2025, Porto, Portugal (AR=29/151)
Right-sizing Recommendations (RSR): Cloud Workload Conformal Prediction for Virtual Machines in Data Center Operations
Proceedings of the 24th IEEE/WIC International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT'25), November 15-18, 2025, London, UK (AR=24.82%).
A Comprehensive Review of Anomaly Detection in Web Logs
Proceedings of the 9th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT’22). IEEE Press, December 6-9, 2022, Vancouver, Washington, USA (AR=(15+5)/55, online)