CV

📄 Download Resume (PDF)

Summary

Senior Platform Engineer building GPU-accelerated Kubernetes platforms for AI/HPC workloads. Specialized in inference observability (vLLM, TTFT, OpenTelemetry) and cluster lifecycle operations across cloud and bare-metal environments. Open-source contributor: Kubernetes, Kueue and KubeAI (vLLM operator).

Experience

SUSE

Senior Platform Engineer · Remote, Italy · Dec 2025 – Present

• vLLM & GenAI Observability: Engineered OpenTelemetry (OTel) connectors to instrument vLLM inference (Time-to-First-Token and related KPIs), enabling on-call triage for multi-tenant GPU inference platform.
• High-performance GPU Monitoring: Engineered GPU observability solution for Kubernetes and KubeVirt clusters, enabling fine-grained monitoring of NVIDIA A2/H100 with MIG and vGPU segmentation, unlocking 40+% HPC efficiency.
• Cluster lifecycle & Autoscaling: Optimized platform efficiency achieving a 62% reduction in infrastructure costs ($100K+ annual savings), by deploying Cluster Autoscaling with Cluster API on multi-cloud (AWS, GCP, On-Prem) platform.
• AI-driven Observability: Architected and delivered the SUSE Observability MCP Server from greenfield idea to MVP, embedding LLM-driven analysis into the alerting pipeline; recognized by senior leadership for roadmap inclusion.
• Infrastructure Automation: Designed and implemented a Kubernetes operator to orchestrate large-scale virtual machine migrations from KVM to Harvester, enabling the migration of 100+ VMs.

Platform Engineer · Remote, Italy · Jan 2025 – Dec 2025

• Infrastructure Governance: Designed and implemented a custom Kubernetes operator to synchronize VLAN and virtual machine assets into NetBox, establishing a single source of truth for infrastructure inventory.
• Federated Observability: Architected migration from Prometheus/Grafana stack to SUSE Observability (StackState) for federated multi-cluster observability, cutting troubleshooting time by 25%.
• Cloud Evolution: Led cloud architecture evolution across AWS, GCP, and Azure for enterprise migrations, aligning Cloud Landing Zone (CLZ) design with strategic requirements, acting as advisor to internal customers.

Stack: Kubernetes, Rancher, Cluster API, KubeVirt, GPU (NVIDIA MIG/vGPU), vLLM, OpenTelemetry, Prometheus, Grafana, Terraform, GitOps (ArgoCD/Flux), AWS/GCP/Azure

Ericsson

DevOps Engineer Team Lead · Pagani, Italy · Sep 2023 – Dec 2024

• Self-Service Platform: Led team of 5 engineers in designing Internal Developer Portal (IDP) based on Spotify Backstage, improving developer efficiency by 25% through self-services. Evangelized portal usage and adoption of best practices.

DevOps Engineer · Jun 2022 – Sep 2023

• ML-driven Optimization: Engineered an ML-driven solution for automated Kubernetes microservice performance optimization, reducing the engineering time required to fine-tune applications configuration from weeks to days.

Cloud Engineer · Mar 2021 – Jun 2022

• Resource Efficiency: Optimized Ericsson Licensing solution footprint by 25% deploying auto-scaling and fine-tuning resources configuration, leading to a throughput increase of 10%.

Stack: Kubernetes/OpenShift, Docker, Helm, Backstage, Gatekeeper (OPA), GitLab CI/Jenkins, Kafka, PostgreSQL/Cassandra, Go, Python, Java

Cisco

ML Engineer · Vimercate, Italy · Apr 2020 – Mar 2021

• Designed and implemented ML-based solution for optical devices troubleshooting, improving defect detection by a 2x factor.

Stack: Python, TensorFlow, Computer Vision

Education

Master Degree in Computer Engineering

Politecnico di Milano · Milan, Italy · 2018 – 2020

Bachelor Degree in Computer Engineering

Università degli Studi di Salerno · Fisciano, Italy · 2014 – 2018

Skills

AI & GPU Infra

NVIDIA MIG/vGPU, GPU-Operator, LLM-Ops, vLLM, Kueue/Slurm, TensorFlow, Pytorch, Computer Vision

Observability

OpenTelemetry (OTel), Prometheus, Grafana, Alertmanager, StackState, Root Cause Analysis (RCA)

Reliability

SLIs/SLOs, alerting strategy, runbooks, incident response, postmortems, capacity planning

Cloud Native

Kubernetes, Helm, Docker, GitOps (ArgoCD, Flux), Terraform, GCP, AWS, Azure

Development

Go, Python, Java, Rust, K8s Operators, Event-Driven Architecture, Linux