AI Test Engineer by BlinqIO
Playwright code + self-healing tests w/ no vendor lock-in
467 followers
Playwright code + self-healing tests w/ no vendor lock-in
467 followers
BlinqIO’s AI Test Engineer builds and maintains Playwright automation in minutes - generating real code in your repo, integrating with CI/CD, and self-healing when your app changes. No lock-in, just faster, stable, zero-touch testing.
Hi Product Hunt,
I’m Tal, founder and CEO of BlinqIO - the AI Test Engineer for end-to-end testing.
We’re engineers ourselves, building for the technical community, and we’ve felt this pain firsthand:
test automation is one of the biggest bottlenecks in software release cycles - especially E2E testing.
We built BlinqIO to remove that bottleneck entirely.
BlinqIO introduces Vibe Testing:
An AI Test Engineer that uses GenAI to create, maintain, execute, and analyze end-to-end tests automatically.
Under the hood, BlinqIO generates real Playwright test code, stored in your own Git repository - so there’s no vendor lock-in. You fully own the tests.
The result: E2E testing goes from a constant human burden to an autonomous AI capability, finally unblocking faster, safer releases.
It is free to get started.
What’s next: we’re already working on extending the AI Test Engineer to support API testing.
We’d love your feedback: What other testing areas or workflows would you like us to support next with the AI Test Engineer?
Tal
@barmeir_tal Hi Tal, this sounds very interesting. congrats on the launch. How do you address test planning with this; bugs vs bad test?
AI Test Engineer by BlinqIO
@barmeir_tal @zolani_matebese Thanks for the question! BlinqIO supports test planning through automated test creation, structured test execution, and detailed failure analysis. When something breaks, the platform provides clear diagnostics and AI-assisted root cause insights so teams can quickly understand what happened, why, and where. This helps streamline the decision process during planning and triage by reducing the time spent digging through logs or guessing at failures.
AI Test Engineer by BlinqIO
@barmeir_tal @zolani_matebese During AI Recovery of a failed test, the AI can determine if the failure is due to a UI change or a problem with the test itself (login credentials wrong, step missing, etc). You can also review the fixes made by the AI Test Engineer before accepting them. You can reject a fix and have a ticket be created in Jira for a bug that AI identified.
@sapnesh_naik Thank you Sapnehs
@barmeir_tal Hey Tal, I have been hunting some products in the testing space, hence I am curious about what parts of a real-world test suite are included by default and what usually still needs human refinement? :)
AI Test Engineer by BlinqIO
@barmeir_tal @rohanrecommends In a real-world BlinqIO suite, the AI Test Engineer automatically generates the automation code that performs complex UI workflows, and maintains selectors dynamically when the application changes.
However, human refinement is still needed to define high-level business logic, validate edge cases, and ensure the "look and feel" aligns with brand-specific UX standards. Essentially, the AI handles the heavy lifting of automation, execution, and maintenance, while the human engineer focuses on strategy and and supervising the AI Test Engineer.
This could save so much time for small teams without dedicated QA! The automated test generation sounds promising. Does it integrate with existing CI/CD pipelines, or does it work as a standalone tool?
AI Test Engineer by BlinqIO
@easytoolsdev BlinqIO works with existing CI/CD pipelines rather than replacing them. The AI generates real Playwright tests that live in your repo, so you can run them in GitHub Actions, Jenkins, GitLab, CircleCI, or whatever you already use. You can also run tests locally during development. So it is not a closed standalone system - it fits into your current workflow and tooling.
AI Test Engineer by BlinqIO
@easytoolsdev Yes. You can run it from your existing CI/CD pipelines. However, we recommend using the BlinqIO Cloud Execution platform included with your license, as it supports parallel execution, AI Recovery, Retrying failed tests, Videos, Traces, and more!
Chatscout by Zevi
Does the AI recommend which tests should be prioritised based on risk or production usage data?
@hiya_chaplot1 Thanks for the question. The AI Test Engineer does not prioritize tests based on risk models or production usage data. Today the focus is on generating, maintaining, and executing real Playwright tests, and letting teams decide prioritization based on their own context. Happy to share more if you are curious.
AI Test Engineer by BlinqIO
@hiya_chaplot1 We also have a new feature coming soon that will let you connect your Jira project to BlinqIO so that the AI Test engineer can suggest test cases for tickets that you select. This will be released in the next two weeks!
Wow, this looks amazing! BlinqIOs AI Test Engineer building Playwright code directly in my repo is a game changer. How well does the self-healing adapt to more complex UI changes, like refactoring a whole component?
AI Test Engineer by BlinqIO
@jaydev13 Thanks! The BlinqIO AI Test Engineer is extremely good at even complex UI changes that cause test failures!
One customer had a test that filled out a large transport order form with multiple sections in a single long page. Later, their product team refactored that flow into a multi-step form wizard, splitting each section into its own step.
And when the test failed, the AI Test Engineer understood that the test scope remained the same even if the UI structure had changed. The AI Recovery automatically generated new Playwright code and locators for the UI and recovered the test without any manual intervention!
AI Test Engineer by BlinqIO
@jaydev13 Thank you! We also belive it's a game changer.Appreciate your feedback.
Prava
From a QA workflow perspective, how does your vision about Vibe Testing change collaboration between developers, QA engineers and product managers over time?
AI Test Engineer by BlinqIO
@zerotox Good question! I think that, with Vibe testing, test maintenance, and deciding what to test/when to test/which tests to run, all of these critical tasks get much easier as AI becomes increasingly impactful. We at BlinqIO think the QA role will evolve into a supervisor role for multiple AI agents that do most of the hands-on work!
FuseBase
Congrats on a long-awaited launch, team @barmeir_tal @liat_niederman @rebecca_villalba @sapnesh_naik
Do tests run locally, or do they go through your cloud? Asking from a security perspective.
Hi Kate
Both options are supported - local execution and cloud execution - and you choose based on your workflow.@liat_niederman @rebecca_villalba @sapnesh_naik @kate_ramakaieva
ThankThank you, excited to finally share it
AI Test Engineer by BlinqIO
@barmeir_tal @liat_niederman @rebecca_villalba @kate_ramakaieva
You have full flexibility in determining where you want your tests to run.
Locally on your laptop
Recommended - On the BlinqIO Execution Cloud (Support parallel runs, AI Recovery, Failed test retries, and videos/traces for failed tests).
External CICD/Servers - You can run it on your own third party CI-CD like Github Actions, Jenkins, Azure, AWS etc or any Windows/Mac/Linux server enironment.
Hope this helps!
Sapnesh
@sapnesh_naik Thank you!
Flexprice
E2E tests being the release bottleneck is painfully accurate. Most teams dread touching them.
Curious how BlinqIO handles flaky tests over time. Is that mostly learned behavior or rule-based?
AI Test Engineer by BlinqIO
@shreya_chaurasia19 You're right! The BlinqIO AI Test Engineer learns from successful test runs and uses that knowledge to recover failed tests.
Each successful run generates a run report that includes screenshots, console logs, network logs, and the Playwright trace file. The AI Test Engineer analyses this and builds a mind map of your application. It learns about your application just like a human engineer would! And this helps it improve over time at recovering from failed tests.
Attaching an image below on how the AI Mind map looks after you have a few tests automated and run.