vikingQA is an AI QA colleague for your whole team. The agent explores, writes the test, runs it on every release and keeps it working.
From a request, a recording, or any change to your product
The agent opens your application and clicks through the flow by itself, the way a tester would.
What it found becomes a test case anyone can read, and an automated test saved with your code. It runs until it passes.
Click through the flow once with the browser extension. The agent turns it into a test case and an automated test.
The agent reads your specs, tickets and past runs, and remembers how your product works.
On vikingQA or on your own infrastructure
Tests run in isolated browsers in the vikingQA cloud, with video and evidence. Your own infrastructure can run them too.
Nightly run · 12 of 20 passed
Run on every deploy or every night. Runs that start when a ticket moves to "Ready for QA" are on our roadmap.
Without a maintenance backlog
When your application changes and a test is out of date, the agent explores again and heals the test. Your team can review the change first, if you want.
Every run comes with a video and the full technical logs. A failed run becomes a bug report in Jira, Linear or Slack.
Session replay
We build vikingQA with a small group of design partners.Join the design partner program
vikingQA keeps one test suite for the whole team. A developer, an analyst and the agent each add to it from their own tools.
One test suite for the whole team. It grows with every release.
Developers and automation engineers add tests as code, from the editor and repository they use now. Tests from the agent arrive in the same repository, in plain Playwright.
For developersBusiness analysts, functional consultants, key users and clients click through a flow once, or describe it in chat. The agent writes the code.
For business teamsThe vikingQA agent writes the test case, explores your application, automates the test, runs it and heals it when your application changes. It adds its work to the suite by itself, or your team reviews it first.
Meet the agentYour tests are stored in your own code repository, in a standard open format.
Your team and the agent build, run and heal your tests here.
test('a customer can place an order', async ({ page }) => {
await page.getByRole('button', { name: 'Place order' }).click();
});If you cancel, every test stays and keeps runningSAML and OIDC, with SCIM provisioning.
Owner, Admin, Member and external Customer roles.
Every action recorded, with export.
Hosted in the EU. A US region is on our roadmap.
Fixed-IP egress per project, so you can allow-list vikingQA. Private runners inside your network are on our roadmap.
Encrypted at rest with AES-256, injected at the moment of use and never sent to the model.
DPA, invoicing and purchase orders. A 99.9% uptime SLA is on our roadmap.
Start free with your own application. No credit card.