Infrastructure and audit
A guarantee is only as real as the machines behind it.
We don't resell a link to someone else's server and hope for the best. Everything we sell runs where we can see it, measure it and fix it.
Where the software runs
Our network. Our machines. Our monitoring.
When a developer joins the catalogue, their product is deployed onto infrastructure we own and operate — not left running wherever it happened to live before.
That is what makes the rest of this site honest. We can promise the software works because we are the ones running it. We can measure uptime because it's our uptime. And when a customer reports a problem, the logs that explain it are already in our hands.
You can't stand behind software you can't see.
The audit process
Five stages in. Re-audited on every release.
Nothing goes on sale because a demo looked good. A product enters the catalogue when it survives all five stages — and every new version goes back through audit before it ships.
STAGE 1
Ownership & licences
Code ownership declared and evidenced; every dependency licence checked.
STAGE 2
Security review
What data it collects, where it stores it, which third parties it calls — answered and verified.
STAGE 3
Deployment
The product is built and deployed onto our staging environment, by us, from the artifacts the developer submits.
STAGE 4
Break testing
We use it, load it and try to break it — the way a real customer eventually will.
STAGE 5
Go-live & watch
Production deployment with monitoring attached from minute one. The audit never really ends.
If it's in the catalogue, it survived all five. Every version, every time.
Monitoring
Measured where it runs, not where it's marketed.
Uptime, error rates and response times are recorded continuously for every product, on the machines that actually serve customers.
The same numbers appear in three places: on our monitoring wall, in the developer's portal, and in the service levels our agreements enforce. Nobody gets a prettier version of the truth.
One set of numbers. Three audiences. Zero flattery.
When a product fails
A written escalation path — not a shrug.
When a product misses its service levels, the developer receives a formal, dated cure notice stating the remedy required and the deadline. Most issues end there — fixed, documented, closed.
If failures repeat and are not remedied, our agreement allows us to step in and continue operating the product ourselves, so paying customers are never stranded. Because the software already runs on our infrastructure, that step-in is a real capability — not a threat we couldn't execute.
That is why the guarantee is real rather than rhetorical.