Software QA

Release testing for the whole product, not only the AI.

Regression, automation, API tests and quality gates for your web product, whether or not it has an AI feature yet. A fixed scope, a report every month, and a team that tests for a living.

What is included

A monthly scope with written limits, so that you know what is covered.

  • Release validation

    Each release is checked against the flows that matter before it goes out, up to an agreed number of releases a month.

  • Critical-path automation

    End-to-end tests in Playwright for the journeys your revenue depends on, kept passing as the product changes.

  • API tests

    Contract and integration tests that catch the breaks behind the user interface.

  • CI quality gates

    Checks in GitHub Actions or GitLab CI that stop a bad build before it reaches staging.

  • Defect triage

    Clear, reproducible reports ranked by impact, so that engineers fix the right thing first.

  • Monthly report

    Coverage, flaky tests, escaped defects and the next priorities, on one page.

Release Risk Audit

Seven business days, one product, fixed price. We look at how you release today and hand you a plan.

  • QA process review

    How work moves from a ticket to production, and where defects get through.

  • Critical user-flow review

    The journeys that matter most, tested by hand and documented.

  • Automation assessment

    Coverage, flakiness and how the suite runs in CI.

  • Production risk

    Escaped defects from the last months, and what they have in common.

  • Action plan

    The top ten issues and a 30-day plan, presented in a readout call.

Questions we get asked

Do you replace our QA team?

Sometimes we are the QA team. Sometimes we add automation and leadership to one that exists. Both work.

Which stacks do you cover?

Web products, with Playwright for end-to-end tests, API tests in the language your team uses, and gates in GitHub Actions or GitLab CI.

Can we combine this with AI evaluation?

Yes. One team, one monthly report and one scope for the AI and the product around it.

AI evaluation & QA
Who owns the tests?

You do. They live in your repository, written so that your engineers can run and extend them.

Tell us how you release today.

A few lines about your product, your release cadence and what breaks. We will suggest where to start.

Founding clients get a lower price on their first project.