QA AI Automation Engineer
Welltory is a health-tech company with over 17 million users that helps people take better care of their health through data analysis. We’re consistently ranked among the top 10 health apps in the United States.
About the role (please read this before the requirements — it matters)
This is an R&D role, and we’re upfront about that from the start. There’s no ready-made test automation framework waiting for you to plug into and write tests on autopilot. Right now, our development has sped up (AI, hiring), while manual QA isn’t scaling at the same pace — it’s become a bottleneck for our entire delivery process. We’re solving this not by growing the team, but by building tools: replacing different stages of testing with AI and automation, one by one, validating each against real outcomes.
That means the stack and processes are still taking shape — and that’s a feature of this role, not a bug. Part of the job is classic mobile test automation. Part of it is AI tools and process automation that go far beyond autotests (for example, we already have an experiment where AI navigates the app on its own, compares screens against Figma pixel-by-pixel, and files bugs automatically). The balance between these areas will shift as experiments either prove themselves or get discarded. If you’re looking for a mature process with clearly defined tracks, this role probably isn’t for you — and that’s completely okay. If you’re excited about building those tracks yourself, keep reading.
One more thing that’s important to understand from the start: at the core of Welltory is health data analysis. Our product processes users’ physiological data and delivers personalized insights. As a result, a significant part of testing isn’t about whether a screen renders correctly — it’s about verifying that calculations are correct and the underlying algorithms produce accurate results.
Requirements
What you’ll definitely need:
- Automation background: you’ve written automated tests before and understand how test automation works — where it creates real value and where it becomes maintenance for maintenance’s sake. The specific stack is secondary.
- An R&D mindset: you’re comfortable with uncertainty. You can take a tool with no proven track record, quickly validate the “works / doesn’t work” hypothesis, document the outcome, and move on instead of waiting for someone to break the task into detailed steps.
- Willingness to work with data (using AI): our product is built around analyzing health data and generating meaningful results for users, so many of our QA tasks involve validating calculation algorithms rather than just UI behavior. You don’t need to be a data scientist or analyst, but you should be comfortable exploring data with the help of AI: comparing calculation results across builds, identifying where numbers started to diverge, and verifying that updated algorithms produce expected outputs. If working with data through AI feels intimidating, this role will likely be challenging.
- A genuine interest in AI tools: you haven’t just “heard about AI” — you actively use it in your work, keep up with new tools, and understand that test automation in 2026 goes far beyond Appium or XCUITest. It includes AI agents, MCP, LLMs, scripts, and automated workflows that glue entire processes together.
- Engineering thinking beyond autotests: you naturally spot repetitive work and think about eliminating it — whether through automated tests, scripts, AI pipelines, or process improvements. We’re optimizing much more than test execution: acceptance testing, regression, traffic analysis, bug workflows, and more.
- Understanding what you’re automating: you have a solid understanding of how a mobile application communicates with the backend (client ↔ server, APIs, state management, analytics). You can’t automate a check for something you don’t understand.
- Async self-sufficiency: we’re a fully remote, asynchronous team. You move your own work forward, ask que