whoami
Computer engineer with a master’s degree in intelligent systems, currently completing a PhD in artificial intelligence.
I design
intelligent components that live inside cybersecurity operations, reading
adversarial signals from noisy systems and rewriting them into actionable intelligence.
I design advanced anomaly-detection systems for network traffic, system logs and connected-vehicle telemetry. The focus is precise identification of behavioral shifts and domain-specific irregularities that signal potential attacks, without relying on stress-induced patterns.
I’ve spent two years researching generative-AI agents and build fully customizable, decoupled multi-agent systems that interact with cybersecurity solutions through natural language. My work covers hallucination-avoidance methods, self-correction mechanisms, and real-time expert-feedback loops that keep reasoning aligned until the target task is completed, reducing the complexity of expert tools for non-expert users.
I build generative-AI systems that produce static and dynamic detection rules for SOC and SIEM platforms using standard, shareable formats. These rules are validated through controlled attack reproductions and static analysis of malware and goodware corpora to ensure low false positives and high family-level coverage. The same controlled environments let me reproduce legitimate behaviors for robust anomaly-detection evaluation.
I maintain independent engineering initiatives: applying AI autonomy to low-cost consumer drones; gaze-driven reading tools; analytics pipelines that derive reading statistics from e-reader annotations; and an AI-based news-analysis project that identified political bias, assessed article quality, and issued trust marks, reaching the regional finals of the Santander X Spain Awards. I also run a startup that digitizes restaurant menus and QR workflows for local businesses, providing detailed interaction analytics and enabling respectful, non-intrusive advertising for nearby businesses with full campaign-visibility metrics.
I design advanced anomaly-detection systems for network traffic, system logs and connected-vehicle telemetry. The focus is precise identification of behavioral shifts and domain-specific irregularities that signal potential attacks, without relying on stress-induced patterns.
I’ve spent two years researching generative-AI agents and build fully customizable, decoupled multi-agent systems that interact with cybersecurity solutions through natural language. My work covers hallucination-avoidance methods, self-correction mechanisms, and real-time expert-feedback loops that keep reasoning aligned until the target task is completed, reducing the complexity of expert tools for non-expert users.
I build generative-AI systems that produce static and dynamic detection rules for SOC and SIEM platforms using standard, shareable formats. These rules are validated through controlled attack reproductions and static analysis of malware and goodware corpora to ensure low false positives and high family-level coverage. The same controlled environments let me reproduce legitimate behaviors for robust anomaly-detection evaluation.
I pursue several independent engineering projects: AI autonomy for low-cost drones, gaze-driven reading tools, reading-analytics pipelines, and an AI news-analysis system that detected bias and evaluated article quality, reaching the regional finals of the Santander X Spain Awards. I also run a startup that digitizes restaurant menus with analytics and privacy-respectful local advertising backed by full campaign-visibility metrics.