# Renato Mignone > I research how AI agents fail and how to make them secure and robust. My work draws on systems security, threat detection, and privacy-preserving authorization. English-language public professional context. Editorial review: 2026-10-30. Professional experience includes security research with the Agentic Systems Lab at ETH Zurich and privacy-preserving authorization research at Huawei Research. Project summaries distinguish academic work, prototypes and demonstrated results. ## Professional context - [Full professional background](https://renatomignone.github.io/profile.md): Start here for the story, experience, direction and evidence boundaries. - [Complete professional context](https://renatomignone.github.io/llms-full.txt): One-file profile with experience, education and all project summaries. - [Homepage](https://renatomignone.github.io/index.md): Short introduction and selected projects. - [Experience](https://renatomignone.github.io/experience/index.md): Public experience records. - [Leadership & Community](https://renatomignone.github.io/leadership-community/index.md): Public leadership & community records. - [Education](https://renatomignone.github.io/education/index.md): Public education records. - [Achievements](https://renatomignone.github.io/achievements/index.md): Public achievements records. - [Projects](https://renatomignone.github.io/projects/index.md): Problem, contribution, results, scope and source links for every project. ## Selected project documentation - [VitaeContext](https://renatomignone.github.io/projects/VitaeContext/index.md): Give agents focused, reusable professional context without copying private files into every project. - [Linux eBPF Verifier Research](https://renatomignone.github.io/projects/Linux-eBPF-Verifier-Bypass-Research/index.md): Understand the gap between verifier acceptance and runtime behavior in Linux eBPF. - [AI-Driven Threat Detection](https://renatomignone.github.io/projects/AI-Driven-Threat-Detection-Research/index.md): Evaluate what ML detectors learn across network, malware, anomaly and shell-command datasets. - [B to Bern (B2Bern)](https://renatomignone.github.io/projects/B-to-Bern/index.md): Connect company needs to relevant public research and expertise in Bern. ## Optional - [From LLMs to Secure Agents](https://renatomignone.github.io/projects/From-LLMs-to-Secure-Agents/index.md): Educational guide in development. - [Sitemap](https://renatomignone.github.io/sitemap.xml): Canonical HTML pages.