01
Git & PR workflow
Branch strategy, branch protection, required reviews, CODEOWNERS, PR and issue templates, commit conventions and versioning.
Services
Delivery & Test Automation, as an ongoing ownership role.
Mekova makes sure every change reaches production safely, tested and predictably — and that it stays that way as the product grows. You choose how much time per week, and for how many months.
What Mekova does
Four parts that together form the route from pull request to production. Every package works on all four.
02
CI and quality gates
Lint, typecheck and unit tests that block a merge as soon as something fails — including build, security check and missing approval.
03
Automated testing
Unit, integration, API, E2E and regression tests. The heart of the service: the suite grows with every new feature.
04
CD and deployment
Build, deploy, verification and rollback. Automatic, with approval, staged, blue/green or canary — depending on the risk.
Growth layers
Within a Partner engagement, these layers follow once the core runs — when you're ready for them. They're not sold separately: they build on a pipeline that's already reliable.
Security automation
Find vulnerabilities and leaked secrets before they reach production.
- Static code analysis (SAST) on every pull request
- Automated dependency updates and dependency review
- Secret scanning with push protection
- Security checks as a quality gate, not a separate report
When: Usually in month 3; earlier with compliance requirements or after an incident.
- CodeQL
- Dependabot
- gitleaks
- SonarQube
Release automation
Releasing becomes a merge, not a ritual.
- Version numbers and changelog generated from commits
- Release notes and tags for every release
- Approval for production where the risk calls for it
- One click or one merge to roll back
When: As soon as you release weekly or more often.
- GitHub Actions
- Azure DevOps
Environments and IaC
Every environment reproducible, defined as code.
- A preview environment for every pull request
- Staging and production as code, not by hand
- Secrets and configuration per environment
- Drift detection: differences between code and reality made visible
When: Often in month 2, once manually managed environments start to chafe.
- Terraform
- Bicep
- Docker
- Kubernetes
- Azure
- Vercel
Observability
Know what happens after a deploy, before a user reports it.
- Health and readiness checks
- A smoke test after every deploy
- Error tracking with deployment markers
- Alerts that reach the right channel
When: As soon as you deploy more than weekly.
Delivery metrics (DORA)
See whether delivery really improves — in numbers, not gut feeling.
- Deployment frequency and lead time for changes
- Change failure rate and time to restore
- Pipeline duration and flaky-test rate
- Discussed in the monthly review
When: From month 4, once the basics are in place and there's something to measure.
- GitHub Actions
- Azure DevOps
Supply chain security
Prove what's in a build and where it came from.
- A software bill of materials (SBOM) with every release
- Signed builds and container images
- Pinned versions of dependencies and CI actions
- Container scans before deployment
When: For clients with security or compliance requirements, or larger organisations.
- Docker
- GitHub Actions
Developer experience
New developers are productive on day one.
- One command to start everything locally
- Devcontainers with the right versions
- Pre-commit hooks for formatting and linting
- Onboarding documentation that's accurate
When: For growing teams, or when onboarding takes days today.
- Docker
- Node.js
- .NET
Tools I work with
Only what I genuinely master. If you use something else, you'll hear honestly whether I'm the right partner — or how quickly I can pick it up.
- CI/CD
- GitHub Actions
- Azure DevOps
- Testing
- Playwright
- Vitest
- Jest
- xUnit
- NUnit
- Cypress
- Postman/Newman
- k6
- Testcontainers
- SonarQube
- Languages and frameworks
- TypeScript
- Node.js
- C# / .NET
- Next.js
- React
- Containers and cloud
- Docker
- Kubernetes
- Azure
- Vercel
- IaC and security
- Terraform
- Bicep
- CodeQL
- Dependabot
- gitleaks
Three packages
You only pay for the hours actually worked, up to a weekly maximum. At the end of each month you get an invoice with a timesheet.
One fixed day a week
Partner · 1 day a week
Up to 8 hours a week · At least 3 months, then as long as you like
€95per hour, excl. VAT
up to ± €3,300 per month
Mekova owns your delivery and test automation and grows it with the product. The main package for teams that keep building.
- The rhythm: core, broaden, harden, own
- Test coverage keeps pace with new features
- Flaky tests, pipeline speed, dependency and security updates
- Weekly update and monthly review
For larger or faster teams
Partner · 1.5 days a week
Up to 12 hours a week · At least 3 months, then as long as you like
€95per hour, excl. VAT
up to ± €4,950 per month
The same ownership, with more room. For teams with several services and environments, a high build pace or a large testing backlog.
- Everything in Partner · 1 day a week
- Room for several services, repositories and environments
- Faster through the rhythm, and earlier into the growth layers
- Weekly update and monthly review
Light follow-up after Partner
Delivery Maintenance
Up to 4 hours a week · Monthly, only after a Partner engagement
€95per hour, excl. VAT
up to ± €1,650 per month
Once delivery is stable: maintenance, updates, monitoring, security and test upkeep. Only as a follow-up to a Partner engagement, so the setup is already known.
- Dependency and security updates
- Test suite upkeep and flaky tests
- Pipeline monitoring and small improvements
- Monthly review
How the arrangements work
- Term
- Partner packages run for at least three months. You decide how many months; after the agreed period we can extend.
- Only hours worked
- You pay €95 per hour for the hours actually worked, up to your package's maximum. If an extra hour is ever needed, we agree on it up front.
- Monthly invoice
- An invoice with a timesheet at the end of each month. Payment within 14 days.
- Scale up or down
- Switching between 8 and 12 hours a week is possible per calendar month, if we agree on it before that month starts.
- Remote, on site by arrangement
- The work happens remotely. For a kick-off or an important moment I'm happy to come by.
- Contact
- Available in your Slack or Teams, a short update every week and a review every month.
How we start
- 1
Introduction
A free 30-minute conversation about your product, team and delivery.
- 2
Scope document
A short document with goals, package, term and arrangements.
- 3
Access
Access to repositories, CI and environments — only what's needed.
- 4
Intake day
The first working day: review the existing pipeline and tests. The outcome is the backlog.
What a Partner engagement demonstrably delivers
- Git
- Branch protection, required checks, PR workflow, CODEOWNERS and templates.
- CI
- Lint, typecheck and unit tests as a blocking gate.
- Testing
- Unit, integration, E2E and regression — growing over time.
- CD
- Build, deploy, rollback and deployment verification.
- Automation
- Dependency and secret scanning and release automation.
- Docs
- Delivery workflow, deployment procedure and onboarding.
When Mekova isn't the right fit
- Projects without ongoing development
- One-off jobs of a few days, like a single GitHub Action or a Dockerfile
- Helpdesk or general IT management
A Partner engagement runs for at least three months, because the value lies in growing with your product.
Are you actively building, and are releases or tests getting harder?
Then this is exactly the right moment. Tell me briefly how you ship today; you'll hear from me within two working days.