Strong Middle AI-Native SDET

We are currently looking for an experienced Strong Middle AI-Native SDET

Mercury Development is a software engineering company specializing in custom solutions for US-based businesses, from large enterprises to fast-growing tech startups. Our apps are used by more than 50 million people worldwide, and many of our products have been featured in the App Store and covered by TechCrunch, Engadget, and Gizmodo. We are a global team of 500+ talented professionals working on a wide range of challenging projects.

We are looking for a Strong Middle AI-Native SDET who uses AI agents at every stage of the task: from requirement analysis and test design to failure analysis and reliable CI runs. The role covers web and mobile applications, APIs, test data, and CI.

The Role

You will independently own automation tasks in an existing project: understand requirements, explore code and solution options together with an AI agent, implement checks and bring them to stable CI runs. The output produced by an agent must be validated just like your own code. Larger changes to the test architecture will be discussed with the team.

Responsibilities

• Design and develop API, UI and E2E tests for web and mobile applications.

• Use AI agents for code analysis, test development and refactoring, preparing test data, diagnosing failures and working with CI.

• Choose the appropriate verification level for each task (API, UI or E2E) and align ambiguous decisions with the team.

• Improve the existing test architecture through small, understandable and verifiable changes.

• Work with test data, CI, test reports and run artifacts.

• Investigate failures, fix causes of instability and maintain coverage as the product changes.

• Participate in code reviews and share solutions with the team.

Requirements

• Practical experience using AI agents in development or automation: ability to define tasks, provide context and constraints, supervise execution and verify results.

• Commercial automation experience with examples of tasks you independently brought to regular CI runs.

• Strong programming skills in TypeScript/JavaScript or Python. Good knowledge of at least one of these languages is required.

• Practical experience with UI/E2E frameworks and API automation.

• Understanding of which level (API, UI or E2E) is best for automating a given check and the ability to explain your choice.

• Experience with Git and CI/CD; ability to diagnose issues from code, logs, reports and artifacts.

• Ability to understand an existing project and its test architecture, decompose tasks and timely flag risks or missing context.

• Ability to validate changes with builds and targeted test runs and prepare them for code review.

• Technical English for reading requirements, documentation and code.

AI as a working tool

AI agents are integrated into everyday work: we use them to explore code, develop and maintain tests, prepare data, analyze logs and CI, and update documentation. For us, an AI-native approach is not about generating a code snippet on request. It is important to decompose the task, provide the agent with the right context, correct its work and verify changes through review, builds and test runs.

Tech stack

• AI tools: Codex, Claude Code, Copilot/Cursor and MCP integrations.

• Other stack depends on the project: TypeScript/JavaScript, Python, Playwright, Detox, Appium, Patrol, Pytest, GitLab CI/CD, Docker, Allure, REST API, SQL, AWS.

Nice to Have

• Experience configuring workflows for AI agents: project instructions, skills, MCP integrations, helper scripts and quality gates.

• Automation of React Native apps with Detox, native iOS/Android apps with XCUITest/Espresso or Flutter apps with Maestro/Patrol.
• Parallel runs, data isolation and test account factories.
• Load testing or visual regression testing.
• Experience with AWS services in test environments.