Case studies / QA flow

Achieving 12x faster test authoring with Al QAagents

5 min

A full QA suite written, against an hour by hand

22 hrs

Back on a QA engineer's calendar every week

~$64k

Capacity returned per engineer, per year

Every ticket starts the same way. Read the user story, think through the edge cases, work out what the change touches elsewhere, write the cases out. On anything non-trivial that is close to an hour of work, and it happens before a single test has been executed.

Then the suite has to be run by hand. Open the case, read step one, go do it, come back, read step two, with the engineer present for all of it. Anything that fails gets written up as a bug, which is a second pass over work already done once. It is careful, necessary work. Almost none of it is creative, and it takes most of the week.

What we built

QA flow runs the front half of the ticket. It reads whatever spec exists, writes the cases a QA engineer would write, runs them unattended, and files what breaks. Three things stay with the engineer: approving the suite, adding the edge cases only they would think of, and fixing the bug. Everything between those decisions happens without anyone watching.

Writes the tests from whatever spec exists

A Jira ticket, a Linear story, a screenshot, or the repo itself. Nothing needs cleaning up first. Take the blend a QA engineer would build, or name the type you want, from end to end down to unit. Review starts from a complete draft, not a blank page.

Runs them without anyone watching

Start the run and walk away. No reading step one, doing it, coming back for step two. Failures become tickets on their own, formatted so a developer can act without a follow up question. The write up step disappears.

Reports what broke, and what keeps breaking

Every run comes back as a summary the team can see. A release reads as a glance, not a question. Failing types surface alongside the counts, and UI breakage looks nothing like unit breakage. Patterns show up across runs, which is usually where the real fix hides.
“Authoring used to take an hour a ticket. Now that hour goes to the edge cases nobody had time to test.”
Disha Rana

Engineering, Islands

Three stages of manual work come off the ticket, and the three decisions worth making stay. What comes back is about twenty two hours a week per engineer, spent on testing that never fit before.




Want this pipeline running on your backlog? Get in touch.

About QA flow
QA flow reads a user story, writes the test cases a QA engineer would write,runs them, and files the bugs it finds. It was built in-house over six months by a full-time pod, and theaccuracy work took longer than the product.
Tags
Product & engineering
Tech stack
Jira
Github
Linear
Want similar results?
We build AI agents and software for teams ready to move faster.
Get in touch
contact image