PhD · AI Systems · HPC + Cloud
I build autonomous AI systems that act on infrastructure — not just explain it.
Platform engineer and researcher at the University of Bologna specialising in MLOps, AIOps, and Kubernetes-native infrastructure. I architect production-grade LLM agent systems, anomaly detection pipelines, and hybrid HPC/cloud orchestration frameworks for datacenter-scale environments.
Seven years of hands-on systems and network engineering in mission-critical industrial environments (power plants, enterprise OT networks) before a PhD in High-Performance Computing gives me a lens most ML researchers lack: I care about uptime, observability, and correctness in production — not just benchmark performance.
My current focus is autonomous infrastructure control: systems that detect failures, reason about root causes, and propose or execute remediation with human approval gates.
LLM-orchestrated multi-agent framework for autonomous Kubernetes operations — root cause analysis, diagnosis, and HITL-gated cluster actions across the full API surface.
Graph anomaly anticipation framework for exascale HPC — topology-aware node failure prediction on Tier-0 CINECA supercomputer (Marconi100).
Thermal hazard prediction framework for datacenters — multi-modal deep learning for sub-100ms thermal failure forecasting at minimal runtime overhead.
Agent Operations Benchmark — evaluates AI agents on HPC operational tasks with role-aware, permission-enforced, trace-based scoring against deterministic environment snapshots. No live cluster required.
CLI and Python SDK for interacting with KubeIntellect from the terminal — streaming responses, Rich TUI, and pipeline-friendly output formats.
PDP 2025 · PDP 2026 · AsHES 2026
IEEE TCAD · FGCS · Journal of Grid Computing · SC · ACM CF · DATE · PDP · AsHES
2 PhD co-advisees (ongoing) · 5 MSc theses · Lab of Big Data Architectures, UniBo (2020–2024)
Open to research collaborations, open-source contributions, and industry partnerships in AI infrastructure and HPC systems.