DevOps Engineer @ ClearML | Observability & Kubernetes (CKAD) | Cloud & MLOps infrastructure

Public link to this page at Filippo Brintazzoli - Resume

<aside> 👋

Product-oriented DevOps engineer specializing in Kubernetes, observability, and GPU scheduling infrastructure for MLOps teams - in the cloud and especially on-premises. I build and scale systems that stay legible and reliable under pressure, focused on keeping ML workloads reliable at scale and cutting teams' time from experiment to production. CKAD and Prometheus certified.

</aside>

Experience 👨🏻‍💻

DevOps Engineer (Freelance)

*ClearMl (MLOps Platform and AI Orchestration Engine), Remote, Italy – (Apr 2024 - Present)*

Infrastructure and platform engineering for ClearML's MLOps platform. Kubernetes focus, reporting to the CTO.

Own deployments for 20+ enterprise on-prem customers end to end: POC design, production rollout, and live technical support. Environments span Kubernetes and bare metal, with fleets of up to hundreds of GPUs.
Design GPU orchestration for environments ranging from single-node bare-metal installs to multi-cluster enterprise systems.
Build internal tooling and observability for managed environments, including self-service isolated deployments.
Maintain ClearML's open-source components.

Observability Tech Lead

Infovista (NLA Cloud Platform), Remote, Italy – (Jan 2023 - Apr 2024)

Delivered Observability-as-a-Service for all of Infovista's teams and products as part of the Platform team: metrics, logs and traces from distributed multi-cluster environments, monitoring thousands of microservices in production.

A single instance could sustain more than 5M active series and 4k logs/second, collected from up to 70 Kubernetes clusters, hundreds of nodes, and more than 5k running pods.
Owned the full stack based on Grafana, Prometheus, Thanos, Loki, Fluent Bit, Fluentd, Tempo, Argo Workflows, MinIO, and more.*

Software Engineer

Empirix/Infovista (NLACP and Ativa), Remote, Italy – (Apr 2020 - Jan 2023)