Why we are building ShipperAG
Software teams ship faster than ever, and AI coding tools make them faster still. Yet QA is usually the first thing squeezed when a release runs late, the requirements live in documents and tickets, and when something breaks after release nobody can show what was checked before it.
We think the answer is not more opinions about quality but evidence: a clear record of what was checked against your own requirements, what failed, what needs a decision and what was not covered.
How ShipperAG works
- Connect a preview or staging build of your product.
- Add context: requirements, design system or tokens, business rules and API docs.
- The coordinator reads the change and picks the specialists it needs.
- Specialists run real checks: a real browser, real API calls and automated accessibility rules.
- The independent verifier re-runs every finding in a fresh session. Only what reproduces counts.
- Results are sealed in a hash-chained evidence ledger, and a person signs off or shares the report with stakeholders.
Every result lands in one of four states: verified, issue found, needs input or not covered. See the animated walkthrough on the homepage and the full team of AI QA specialists.
What we will not trade for speed
- Your intent defines correct. The code is not its own specification; what you asked for is.
- Autonomous, with evidence. It plans and runs the checks on its own and asks only when a real decision is needed. Nothing is released on trust, and an AI model's opinion is never treated as proof.
- Nothing approves its own work. Whoever changes the code is never the only one who checks it.
- Your code stays yours. Checks are designed to run in your own pipeline, next to your code.
Standards we check against
- WCAG 2.2, the W3C's Web Content Accessibility Guidelines, for accessibility checks.
- The European Accessibility Act, Directive (EU) 2019/882, which applies to many consumer digital services from 28 June 2025. See our accessibility testing guide.
- OWASP Top 10 (top10.owasp.org): security checks target broken access control, the top risk in the list, plus session-cookie and header basics.
- Hash-chained evidence ledger: every check and finding is recorded in order, so a report reflects what actually ran.
How we handle data
- Waitlist: we store what you type into the form and the page you joined from, privately, and use it only to contact you about the pilot.
- Visit counting without cookies: the site sets no cookies and stores no IP address. Visits are counted with a salted hash that changes every day; this is an approximate daily count, not a count of identified people. Browsers that send "Do Not Track" or Global Privacy Control are not counted at all.
- Page interaction measurement: we measure coarse click positions outside forms, scroll reach and form stages to improve the website. We do not record sessions or collect typed values through this measurement. A random ID lasts for one page load; if you join the waitlist, it links that page load to the saved entry for conversion counting. Behaviour events and this link are cleared after 30 days when new behaviour is received. Processing uses our Supabase service. Do Not Track and Global Privacy Control stop collection;
?analytics=offalso disables it for that page. - Fonts and scripts are served from shipperag.com itself, so visiting the site sends nothing to third-party font or advertising services.
- Testing targets: ShipperAG only tests products you own or have written permission to test, and stays inside the hosts you allow.
The company
ShipperAG is an early-stage company run by its two founders. The product is in a private pilot, and we are opening it to a small number of teams from the waitlist. We have not announced general availability yet. Every team starts with a free 45-day trial with up to 20 release checks, and paid plans start at $149 a month; we would rather show you the product on your own work than make claims about it here.
The best way to reach us is to join the waitlist and tell us what you would like to test first. We read every entry. We also post on LinkedIn.