Testifly launch
AI-powered end-to-end full web testing in minutes
362 followers
AI-powered end-to-end full web testing in minutes
362 followers
From URL to fully automated end-to-end web testing in minutes! Testifly makes end-to-end web testing fast, stable, and low-maintenance. After providing your URL and logins, Testifly creates reliable tests in minutes using AI that understands real user flows and adapts to UI changes. No long setup, no brittle scripts, fewer flaky failures. Built on modern browser automation and designed for real production apps, CI pipelines, and fast-moving teams.
This is the 2nd launch from Testifly launch. View more
Testifly
Launching today
CTOs and founders with small teams and no time or resources to test and maintain. Connect your web app and Testifly does the rest. Discovers flows, writes tests, self-heals, explains failures with video. Zero prompting. Zero maintenance. 30-day free trial.
Free Options
Launch Team / Built With
Atlassian LoomTalk it through in Loom. Ship with AI
Promoted
congrats on the launch amihay. biggest pain for me is signup breaking after a small copy or pricing change, nobody tests it because nobody calls it a code change. if testifly catches those too it is already worth it
Thank you @agshin_rajabov ! Yes, it catches those. Testifly checks what a real user experiences in the browser, so it doesn't matter whether the change was code, copy or pricing. If signup breaks after a pricing edit, the signup test fails and you get a video and a root-cause report. With scheduled runs and Slack alerts, you find out within hours and not when a customer complains.
Keeping a history of what a test used to expect would matter here. It would give the team a way to check that a real regression hasn't become the new passing result.
The concern is fair @oleksii_sekundant. A tool that heals itself could, in theory, accept a broken result as the new normal.
Testifly's design guards against this. When a test fails, it separates a UI change (a moved button, a new selector) from a real behavior change (a wrong checkout total). It heals only the first kind. The second kind is flagged as a regression and stays red until someone looks at it. Each run also keeps its video and root-cause report, so you can see what happened and when.