At Toloka, I review and author test cases for a production coding-agent evaluation benchmark. I wrote around 70 test cases end-to-end, including requirements and functional and style validators in pytest and Playwright.
I also write and review tests for FastAPI REST endpoints, checking both responses and stored data. I’ve caught validators that could inflate agent scores and audited LLM-as-a-judge results against agent transcripts and code diffs.
Previously, I wrote automated web tests in Python with Selenium at SecureDen SRL and tested restaurant software on physical hardware at ACROM. I also crowdtested web and mobile apps with Testbirds, and hold the ISTQB Foundation Level certification.
