Honoring #IconsOfQuality: Gleb Bahmutov, Senior Director Of Engineering

To celebrate the relentless passion of testers, BrowserStack is honoring some of the icons in the testing space—those who push the industry forward by sharing their knowledge with the community at large through their thought leadership.

Gleb Bahmutov is a Senior Director Of Engineering at Mercari US. With a passion for software quality and a deep commitment to efficient testing practices, Gleb’s insights are invaluable for anyone navigating the world of software testing and engineering.

Tell us a bit about your role as the Senior Director Of Engineering at Mercari US.

Mercari US is an online marketplace where people list their items for sale and buy things from other people. We only earn money if our mobile application and website work. Quality is important, and I am in charge of QA department which works closely with the engineers to make end-to-end testing easy and efficient.

What’s one piece of advice you would give to someone just starting their career in software development?

Make your learning journey public. Write blog posts or notes for yourself, keep them searchable, and maintain public GitHub repositories with your experiments. This approach not only accelerates your learning but also serves as proof of your knowledge and experience during job interviews.

How do you keep up with all the new trends and tools in software development?

I stay updated by subscribing to several newsletters and regularly reading interesting blog posts. A few newsletters I recommend are The Smashing Newsletter, Node Weekly, and JavaScript Weekly.

What’s a testing trend that’s got you excited these days?

I am excited about running React / Vue / etc component tests in the real browsers instead of JSDom. For too long, the tests were hard to write, nearly impossible to debug, and the code would not work the same way since we were not using the real browser to execute them. Now with Cypress component testing and Vitest browser mode, we finally can launch a browser, "mount" a component, and see it in real action.

Give your users a seamless experience by testing on 20,000+ real devices with BrowserStack.

Has AI changed the way you and your team approach testing? If so, how?

We use AI to help write tests only in small syntax bits via GitHub Copilot. I still have not seen a useful AI system that can write the entire realistic E2E test.

Learn how AI-powered test cases can enhance your testing efficiency.

Looking back, what's the biggest shift your team has made in testing, and what impact did it have?

One of the most significant changes was shifting the responsibility for automated end-to-end tests from QA to developers, at least for the initial review of pull requests. This allows developers to ensure new features don’t break existing ones before QA even steps in, streamlining our entire workflow.

What are your go-to productivity hacks that help you stay on top of your game?

I really really _really_ love small pull requests. If a PR is open for longer than a day, it's a red flag for me. My advice? Split it up, merge the parts that are ready, and keep things moving. Don’t let small, useful changes get blocked because one part of the work is complicated.

(Responses have been edited for clarity.)