QA.tech - Ship faster with AI agents that create and run your QA tests

Hey Product Hunt! 👋 Patrick here, CPO & co-founder at QA.tech. From my experience, building countless web apps, the features we have built have required tons of testing before being shippable. Automated tests have not solved this nightmare for me - apart from taking time to set up, general flakiness, and inability to handle changes - it depends on you to come up with the test cases and make the right checks. I wanted something that could help me create a test plan by just browsing my website or web app, and that detects bugs and improvements at every interaction while running the tests. That’s why we built QA.tech, which uses AI to scan your website or web app and generate a comprehensive test suite that can be run using a swarm of agents as part of your CI pipeline or at your desired frequency without the need for a manual setup. What makes QA.tech stand out? ✅ Time-Saving: Generate full test suites in minutes, not days. ✅ Reliability: Run tests automatically on every release or whenever you need. ✅ Coverage: Our agents detect and create tests for new functionality. ✅ Resilient: Our agents adapt to changes in functionality. Who is QA.tech for? From startups to enterprises, QA.tech adapts to your needs. 🚀 CTOs: Accelerate time to market with faster, automated QA. 🛠️ Developers: Catch more bugs, test your releases, and get your own QA. 🤖 QA Teams: Automate repetitive tasks and focus on higher-value testing. We’ve poured our hearts into this product and are thrilled to share it with the PH community. Extra shoutout to the amazing team that has made this possible. Every new product is a journey we love to hear from builders on how we can shape the best possible product for testing. We know every great product starts with help from the community. Let us know your thoughts and feel free to ask us anything! 🚀

Incredibly smart team and a really cool product! I love that it takes a lot of the tedious work off your plate but gives you complete control to edit and tweak tests when needed. That's how every devtool should work!

Hi there! Vilhelm here! Im one of the Co-founders and CAIO at QA.tech! Im incredibly excited and proud of what we have built here! We have such a cool system in order to deliver great autonomous agents. Everything from custom fine-tuned models and embeddings to pre-process interfaces to mapping the pages into a graph representation making it possible for us to do advanced planning and analysis of what can be tested and continuously learn and adapt to your specific application. I encourage you to give it a go!

Hi Product Hunt! 👋 I'm Alex, an AI engineer at QA Tech. We’re excited to introduce our AI agent that can test web pages for bugs. Building such an agent requires developing a large number of components, including grounding capabilities, planning mechanisms, and state/action representations. The backbone of our agent lies in learning graph representations for each customer’s web page. These graphs capture the structure of the site, representing its pages and all possible interactable elements (actions) available in each state. A key development in our work has been creating unique and consistent representations for these actions. To achieve this, we train Language and Vision Transformers on our internally collected dataset, which comprises page screenshots and their corresponding HTML. For planning, we use a combination of model-based planning, where we simulate interactions using a world model, and Monte Carlo Tree Search applied to our constructed graphs. As a result, our agent continuously improves with use: the more tests it performs on a website, the better it becomes at generalizing and planning for new test scenarios. Try it out and let us know what you think!

Patrick & team have been at the forefront of doing deep innovation in automated testing, and I'd bet on them to build a solution that *really* works. Congratulations on the launch!

Hi everyone! Building the infrastructure for QA.tech has been quite the ride. My days are spent juggling Pulumi, Vercel, AWS, GCP, and Supabase with tools like BullMQ, Inngest and other to create the foundation that lets our AI agents do their magic. Watching AI agents actually run tests that used to take our team days to write still feels like science fiction sometimes - especially knowing all the complex systems under the hood, from graph representations of web pages to fine-tuned models that understand UI elements and user flows! Happy to geek out about the tech stack if anyone's curious! 🚀

Hi everyone! 👋 I'm part of the GTM team here at QA.tech. I wasn't planning on joining another startup but having worked with Daniel and Patrick in the past + falling in love with product I couldn't help myself 😄 To my non-dev friends I like to describe QA.tech as replicating the work of a human QA team using AI agents. It's really like having a human QA that: 👩‍💻 accesses your product (usually a staging environment) 🧪 test all your critical flows 👩‍🎨 use creativity to suggest new tests 🦾 executes the tests at whatever frequency you want 📊 gives you detailed reports and recordings to help you fix bugs fast QA.tech is removing a lot of the pains with test automation and is significantly more cost effective than human QA. Magical? Yes – that's why I had to join 😄 Would love to get your thoughts, feedback, ideas and support!

the bane of every developer... testing! Nice product, Patrick! Are you using your own AI-models or can external ones be used (some companies can't risk data-sharing outside the organisation)?

Hello, I’m Hanadi! 👋 I work with Talent and People at QA.tech, and I’m proud to be part of a company with a mission to simplify the workday for developers. It’s amazing to see what our team has accomplished in just one year, Seeing the impact we’re making while pushing the boundaries of innovation inspires me every day! 🚀

Looks sick! What kinds of products does QA.tech work the best on right now?