Filip Mrhálek — independent platform engineer · est. 2020
$ whoami
I build cloud infrastructure that product teams stop noticing. Azure-leaning, Terraform-shaped, contracting out of Prague — currently with Packeta, previously ČEZ, Lundegaard, BOOTIQ.
06
years contracting
04
long engagements
30+
terraform modules in prod
[ past engagements ]
Independent platform / DevOps engineer based in Prague. Six years contracting for Czech enterprises — currently building Azure infrastructure at Packeta, previously DevOps at ČEZ, Lundegaard, and BOOTIQ.
I write a lot of Terraform. An opinionated library of 30+ Azure modules now powers ~50 product projects on the same conventions — hierarchical tag inheritance, leaf-level backends, every component scaffolded the same way. Code I'd want to read at 2 a.m. on an incident call.
Day-to-day I move between AKS, App Service, Azure SQL, KeyVault, Service Bus, GitLab CI runners, and the occasional foray into Ansible or ClickHouse. I care more about how a system feels to operate than how clever it looks on a slide.
Capacity
2 clients at a time
Length
3–12 months typical
Location
Hybrid from Prague
Rate
On request
Designing and running Azure (and AWS) at production scale
End-to-end platform design: networking, identity, observability, cost guardrails. From a single greenfield project to multi-tenant platforms supporting dozens of product teams.
Opinionated, reusable infrastructure that scales with the team
I write Terraform like other people write essays. Module conventions, hierarchical tagging, scaffold-and-go project layouts. Engineers spin up new infrastructure in minutes instead of days — and code I'd want to read on an incident call.
From pull request to production with proper guardrails
Pipeline design across GitLab CI, GitHub Actions, Jenkins, and CircleCI. Secret hygiene, ephemeral preview environments, ArgoCD-driven GitOps, self-hosted runners that don't fall over on a Monday morning.
The boring helpers that make a team feel fast
Internal CLIs, scaffolding tools, golden-path templates, ergonomic auth flows. Less typing, fewer footguns, more time for actual product work. Some of these get released; most just quietly save the team a few hours every week.
01
~1 week
A short paid scoping engagement. I poke around your repos, talk to your team, write up what's actually going on and what to fix first.
02
3–12 months
Heads-down delivery. Modules, pipelines, platform fixes — whatever the discovery surfaced. Weekly check-ins, monthly retros, no surprises.
03
2 weeks
Documentation that someone other than me can actually use. A trained-up internal owner. A boring exit, the way it should be.
Got something gnarly to build? I have one spot opening soon.
Start a conversation2025 — now
Packeta Group
Logistics · E-commerce
Built and maintained an internal Terraform module library and IaC scaffold conventions used across the platform. Hierarchical tag inheritance, leaf-level backends, consistent project layouts. New infrastructure spins up in minutes instead of days.
30+
Terraform modules
50+
Product projects
1 yr+
Active
2024 — 2025
ČEZ
Energy · Enterprise
DevOps capacity for SAP Fiori and Java / JavaScript application delivery. Jenkins pipelines, Git workflows, and the boring release engineering work that keeps an enterprise unblocked.
14 mo
Engagement
JVM + JS
Stacks shipped
Hybrid
Cadence
2019 — 2023
Lundegaard · BOOTIQ
Agency · SaaS delivery
Where the Terraform habit started. DevOps capacity rotated across multiple client teams — AWS infrastructure, CircleCI pipelines, Heroku deployments, Docker. Fast iteration, many simultaneous projects.
4+ yrs
Combined
Multi-client
Cadence
Cloud-first
Stack
“A short, specific quote from someone who hired Filip — e.g. an engineering manager or CTO. Two or three lines, concrete outcome, real voice. Replace this block when ready.”
“A short, specific quote from someone who hired Filip — e.g. an engineering manager or CTO. Two or three lines, concrete outcome, real voice. Replace this block when ready.”
2025 — now
1 yr 3 mo
Cloud Engineer
Contractor · živnostník · hybrid
2024 — 2025
1 yr 2 mo
DevOps Engineer
Contractor · part-time · hybrid
2022 — 2023
1 yr 1 mo
DevOps Specialist
Contractor · part-time · on-site
2019 — 2022
3 yrs 6 mo
DevOps Specialist
Contractor · part-time
Education
On request — depends on scope, length, and tech stack. For a clean answer, drop me an email with a one-paragraph project description and I'll come back with a number.
Generally no — I aim for 3+ months. Exceptions: a clear-scope discovery week (~1 wk paid) or a defined deliverable like a Terraform module audit (~2 wks).
IČO / živnostník invoicing. I'm a registered Czech sole trader (IČO 09535560, VAT-registered). For EU clients this is the simplest path; for non-EU we can talk.
Hybrid out of Prague is the default. Occasional on-site for kickoffs or workshops. Fully remote is possible for the right team and timezone.
Right now, no — contractor capacity only. I'm happy with multi-day-per-week commitments and long-running engagements, but not headcount.