Developing a machine-learning pipeline for detecting malicious traffic and cyberattacks in Software-Defined Networks.

• Analysing network-traffic features associated with DoS/DDoS and other SDN threats.
• Evaluating Random Forest performance using classification metrics and adversarial perturbation scenarios.
• Investigating model robustness and mitigation approaches such as adversarial training and ensemble models.

Developing a modular personal AI assistant supporting both offline local language models and cloud-based AI providers.

• Implemented tool-calling workflows for web research, file operations, Python execution and terminal commands.
• Added confirmation controls for sensitive system actions.
• Developed persistent Markdown-based memory with keyword and semantic retrieval.
• Designed support for Ollama, LM Studio and cloud LLM APIs.

Developed a Greek and English web application for analysing AI-style writing patterns and document similarity in PDF and DOCX files.

• Engineered interpretable linguistic and statistical features for sentence variation, phrase repetition and writing-pattern analysis.
• Implemented section-level document analysis, suspicious-passage highlighting and revision recommendations.
• Integrated academic and web-source search functionality for similarity investigation.

Developed core components of an AI-assisted Security Operations Center dashboard for monitoring, correlating and explaining network-security events.

• Implemented simulated event collection from Cisco IOS, Cisco ASA, pfSense and IDS environments.
• Developed real-time security KPIs and attack-scenario simulations including port scanning, brute-force attacks, DDoS and data exfiltration.
• Built backend functionality for anomaly detection, event storage, incident investigation and controlled IP-blocking workflows.
• Integrated a natural-language SOC assistant for alert analysis, event correlation and remediation recommendations.

Contributed to the development of an electric racing motorcycle for the MotoStudent International Competition.

• Supported electrical-system development, component integration and vehicle assembly.
• Worked with embedded systems, Arduino-based components and technical data collection.
• Participated in troubleshooting, testing and performance evaluation.
• Collaborated with a multidisciplinary engineering team during design and manufacturing activities.