Our product
Qualyn: the release call for engineering teams
Qualyn reads the signals your pipeline already produces — pull requests, test results, coverage, security findings — and turns them into one explainable release confidence score, so “are we good to ship?” stops being answered by whoever sounds most confident.
Why we built it
Years of release calls taught me how the question actually gets answered. Five minutes before the go/no-go, someone opens the CI dashboard: 1,400 tests, four red. Are those four the usual flaky suspects or something real? Nobody is entirely sure, the deadline is tonight, and the call gets made on gut feeling and seniority. I have been the person saying “ship it” with less evidence than I would accept in a bug report.
Qualyn is our answer to that meeting. It connects to GitHub and reads what your pipeline already produces: pull requests, test results, coverage, security findings. It weighs them into a release confidence score from 0 to 100 and, more importantly, shows its working: which signals moved the score, which failed test is blocking, what would raise it. The recommendation is one of three words: ship, review or hold. Every call comes with reasons, because a number nobody can interrogate is just gut feeling with a decimal point.
The part I would have wanted years ago is the AI-code risk detection. Code generated by Cursor or Copilot merges faster than anyone can properly review it, and it fails in a recognisable way: the happy path works while validation and access control are quietly missing. Qualyn measures how much of a release is generated code with thin test coverage, so at least you know where the ice is thinnest before you step on it.
One honest limit: a score is not a tester. Qualyn tells you where the risk sits; it will not find the bug for you. Someone still has to do the testing, and that someone can be your team or the testology side of ours — we built Qualyn because scoring risk and testing against it are two halves of the same job.
How it works
Connect
Link your GitHub repositories and the test signals your pipeline already produces. No new dashboards to feed, no process to adopt first.
Analyse
Qualyn weighs the change itself: what the pull requests touch, what the tests cover, what the security scan flagged and how much of the diff is generated code.
Score
Everything lands in one release confidence score from 0 to 100, with the working shown: which signals moved it and what would raise it.
Decide
The recommendation is one of three words: ship, review or hold, each with the specific actions behind it, like targeted regression testing on the riskiest flows.
Who it helps
For founders
You're shipping fast, possibly with AI tools writing much of the code, and you can't personally review every change. Qualyn gives you a number you can ask about in one glance and reasons you can read without being the one who wrote the tests.
For CTOs
The release call carries your name. Qualyn turns it from seniority-versus-gut-feeling into an evidence-backed decision with an audit trail, and calibrates its scoring against what actually happened after each release.
For engineering teams
Quality gates live where you work: in the GitHub flow. Failed tests and blocking risks get flagged on the release, not discovered in the go/no-go meeting, and the recurring argument about which red tests matter finally has a referee.
What's inside
Release confidence score
A 0–100 score combining test results, coverage, change risk, security findings and AI-code exposure into one number per release.
Explainable insights
Every call comes with reasons: which signals moved the score, why it changed since the last release, and what would improve it.
Quality gates
Failed tests and blocking risks flagged before the release call, so nothing rides along unnoticed because the suite was 'mostly green'.
AI-code risk detection
Measures how much of a release is generated code with thin coverage, the code most likely to skip validation and access control.
Manual risk register
Some blockers live in people's heads, not pipelines. Log them by hand and they hold the score down until someone clears them.
Operational readiness
Deployment, monitoring and incident-response checks tracked alongside code risk, because shipping is more than merging.
Pricing
Straightforward tiers, as listed on qualyn.ai. Start free on one repository and upgrade when the score earns its seat in your release call.
Developer
Free
1 repository, 25 scans a month. Enough to score a side project or trial it on your riskiest repo.
Team
£199/mo
5 repositories, 250 scans a month, AI insights included. The tier most startups land on.
Engineering
£599/mo
20 repositories, 2,000 scans a month and custom scoring tuned to how your organisation weighs risk.
See your next release's score
Connect a repository and Qualyn scores your next release in minutes. And if the score points at risks you would rather not test alone, you know where we are.