Show HN: QA.tech – End-to-End Testing in Natural Language

qa.tech

7 points by daniel_mp a day ago

Hi, I'm Daniel, CEO of QA.tech. We've spent the past 18 months developing a cool tool that allows you to run end-to-end tests on any web application using natural language. Check it out and let us know what you think!

I used to be a CTO at a B2C marketplace and was taken aback that all qa agencies we talked to wanted to create a custom suite of tests from scratch. Even though we were essentially a e-com with the same functionality as everyone else. Why weren't there any best practices?

Some time later my co-founders and I had moved on and were looking for the next thing to do. AI agents became a thing, and we wanted to use them to take test cases, written by humans, and perform them with bots.

A test case can be very vague, and test the overall functionality, like this: Sign up for an account and send an invoice

Or include as many specific instructions as you choose, like After signup, verify that the email contains an image of the Chinese zodiac of the year you entered as your birth date

The tech is still very early, and a bit slow in some cases, but I am genuinely impressed by how much the agent manages to solve by itself.

Some cool things about the tool:

All tests are recorded and you get a step-by-step replay of what happened

It tracks all console logs and network logs for easier debugging

It creates “Issues” of bugs found during test execution that you can export to Linear

It suggests test cases for you based on what we scrape from your URL

We have an open signup where you can try it out and create your own test cases.

https://app.qa.tech/auth/sign-up?code=site-signup

Hope this brings you some joy, and that you have some feedback for me.

Cheers

If you need a site to test on, I have built a demo site that you can use: https://bit.ly/qa-demo-app

armcat 17 hours ago

This is brilliant! I spent several years with in QA and even built an ML system for categorizing faults. It's incredible how far we've come though, well done!

Do you mind telling us a bit about the AI tech/stack that is used under the hood?