Open to a full-time staff role · Berlin
Staff DevOps / SRE engineer moving into platform engineering
Fifteen years in operations, most recently as Staff DevOps Engineer at LiquidM / Equativ and Senior DevOps Engineer at Everex. I build and run infrastructure from bare metal to Kubernetes: Linux, Terraform and Ansible, operators in Go, PostgreSQL and ClickHouse.
Self-employed alongside my full-time roles since 2020 and fully independent since February 2026, I am now looking for a full-time staff role again.
What I’m looking for
- Role
- A full-time, hands-on staff role where I own the platform.
- Where
- Berlin or remote (EU).
- Start
- Immediately.
- Consulting
- Runs today; whether some of it continues alongside a full-time role depends on the employment contract.
- Direction
- Platform engineering, self-service infrastructure, and AI-assisted operations instead of manual DevOps.
Experience
My titles have been DevOps and SRE; the short version, newest first.
- 02/2020 – today Self-employed DevOps engineer & IT consultant · zhurko.de
Infrastructure for Turbologo, PostPost, Thryve and Go X. Alongside full-time roles since 2020, during Everex mostly occasional incident support for Turbologo; fully independent since 02/2026.
- 10/2023 – 02/2026 Senior DevOps Engineer · Everex
Multi-chain deployment with Kubernetes and custom operators in Go.
- 01/2020 – 10/2023 Staff DevOps Engineer · LiquidM / Equativ
The hybrid bare-metal and cloud platform behind up to 150 billion requests a day and petabytes of analytics data; the move from Chef to Kubernetes. Set the technical direction, ran design reviews and mentored engineers.
- 10/2017 – 12/2019 Senior DevOps / SRE · Free2Move
Moved AWS ECS workloads to Kubernetes; platform upgrades with zero downtime.
- 09/2015 – 09/2017 Senior DevOps Engineer · Virool
With a small team, modernized the legacy infrastructure of an RTB system and cut its costs by more than 50%.
- 2011 – 2015 DevOps and SRE roles
Key achievements
- A platform at 150 billion requests a day
At LiquidM / Equativ, the hybrid bare-metal and cloud platform taking up to 150 billion requests a day, with petabytes of analytics data behind it.
- Costs cut by more than half
At Virool, with a small team, modernized the legacy infrastructure of an RTB system and cut its costs by more than 50%. It stayed reliable for years after active maintenance ended.
- From Chef to Kubernetes
At LiquidM, the move from legacy Chef-managed infrastructure to Kubernetes, on bare metal and in the cloud.
- A new chain as a deployment
At Everex, an internal platform for multi-chain deployment, built on Kubernetes and custom operators in Go.
- Upgrades without downtime
At Free2Move, moved AWS ECS workloads to Kubernetes and brought downtime from platform upgrades to zero.
Projects
What I build on my own: an open-source provisioner, and research into operations that an AI does under supervision.
UMP — Unus Mundus Provisioner
One operating system, in the same shape, on any server: a CLI that installs the same base system, with the same disks, users, keys and firewall, on dedicated servers and cloud VMs at five providers, with no control plane. Open source under Apache-2.0, in Go.
ump.zhurko.de →
AI-assisted operations (R&D)
A Telegram ops bot on Claude tool use: the commands it may run on its own come from an allowlist in YAML, kept in git; harder cases escalate to a stronger model.
Now designing a multi-agent system with MCP servers for Kubernetes, SSH, Ansible, Terraform and monitoring.
Get in touch
E-mail is the quickest way; I reply in English or German.
Looking for a consultant rather than a hire? See what I do as a freelancer.