A few years ago, a tester received a build, tested it and gave a verdict. Today code goes through a pipeline that builds, tests and ships on every change. A tester who cannot read that pipeline only sees part of the work. Here is what recent US-based studies say about it, with their limits.
TL;DR
- Testing moves into the pipeline: it runs on every commit, not once at the end of a project.
- Containers are mainstream: 82% of container users run Kubernetes in production (CNCF, January 2026).
- The main obstacle is human: the most-cited challenge is cultural change with development teams (47%), not technology.
- No figure measures "the market for DevOps testers", and none is claimed here.
What changes: testing is no longer "the last step"
In a team that ships often, nobody waits for a tester to open a spreadsheet of test cases. Automated checks start on every code change, and the pipeline stops when a test fails. The tester gains a more visible role: deciding which tests run, when, in which environment, and how to read a failure.
Employers in France already ask for it. In our QA skills barometer of 2 October 2026, CI/CD appears in 35.1% of the 405 QA job ads we analysed. That is a French sample, not a US one, but it points the same way. For concrete examples with Jenkins, GitLab CI and GitHub Actions, see test automation in CI/CD.
What the US-facing data says, and its limits
Each of these figures measures something specific.
- Kubernetes in production. In the CNCF Annual Cloud Native Survey 2025, published on January 20, 2026, 82% of container users run Kubernetes in production, up from 66% in 2023. In the same survey, 98% of organisations say they have adopted cloud native techniques and 59% say much or nearly all of their development and deployment is cloud native.
- A cultural obstacle first. The same survey names "cultural changes with the development team" as the top challenge (47%). CNCF says it is the first time the main challenge is organisational rather than technical. For a tester, being able to work with developers weighs as much as knowing the tools.
- AI speeds up code, not trust. The 2025 DORA report from Google Cloud (September 23, 2025, nearly 5,000 technology professionals surveyed) says AI adoption among software professionals reached 90%, and is now linked to higher software delivery throughput. It also notes that the ongoing challenge is making sure software works as intended before it reaches users. That is exactly where testing sits.
- Jobs. The US Bureau of Labor Statistics projects employment of software developers, QA analysts and testers to grow 10% from 2025 to 2035. For the QA analysts and testers line alone, it counts 187,600 jobs in 2025 and projects 198,200 in 2035, a 6% increase. The BLS groups these roles with developers for pay and does not break out DevOps testing, so I quote no salary here.
CI/CD, Docker, Kubernetes: seen from the test side
Four simple ideas, without needless jargon.
CI/CD: tests on every commit. A pipeline runs tests automatically when a change arrives. The tester chooses what runs on every commit (fast), on every merge (broader) and overnight (long). A flaky test that fails the pipeline for no reason costs the whole team its trust, so the job is as much about making tests reliable as automating them. If that sounds familiar, our earlier piece on flaky Playwright tests goes through the usual causes.
Docker: the same environment everywhere. A container packages the application and its dependencies. For a tester it ends "it works on my machine": you start the same environment locally and in the pipeline, with controlled test data.
Kubernetes: knowing where tests run. When the application runs on Kubernetes, test environments are created and destroyed there. You do not need to administer a cluster. You need to understand enough to run your tests in the right place and to read a log when something does not start.
Observability: seeing what happens after release. Logs, metrics and traces help explain a failure and spot a regression the tests missed.
A practical first step
If your tests only run on your laptop, move one smoke suite into a pipeline and make it block the merge when it fails. That single change forces every question that matters: where does the environment come from, where does the test data come from, and what happens when a test is flaky. If you want a second pair of eyes on how your tests fit into delivery, ADC also does a Cypress to Playwright migration with both suites running in CI during the transition.
The DevOps QA specialisation, briefly
ADC's DevOps QA specialisation covers CI/CD, Docker, Kubernetes and observability. The programme page lists 399 hours over 3 months, 100% online, with one-on-one mentoring, 6 months on the learning platform and 24 months of replays. The next session is announced for December; exact dates are confirmed with an advisor.
Two honest points. The programme pages state that ADC courses are taught in French, so ask an advisor about the language before you decide. And no certification is promised for this specialisation. Launch price: €1,500 for individuals (1 to 4 interest-free instalments), €7,500 for companies. Public funding schemes are not yet available at ADC. No programme guarantees a job or a salary.
Over to you
Where does your test suite run today: on every commit, on merge, nightly, or only when someone remembers? And in your team, is the harder part the tooling or, as CNCF's 47% suggests, the conversation with developers?
Sources: CNCF: Annual Cloud Native Survey 2025 (January 20, 2026) · Google: 2025 DORA report (September 23, 2025) · US Bureau of Labor Statistics: Software Developers, Quality Assurance Analysts, and Testers (projections 2025-35, page updated August 27, 2026) · ADC: QA Skills Barometer France (October 2, 2026)
Originally published on AutomationDataCamp. Written with AI assistance. Every figure comes from the sources listed above, as cited in the original article (checked on 19 October 2026).
AutomationDataCamp is an online software testing academy (Playwright, API testing, CI, AI-assisted testing, ISTQB prep). More on our site.
Top comments (0)