Andrey Vishin

Java, Python, TypeScript, React, TestNG, Selenium, Appium, Playwright, Firebase, LLM output evaluation

Evaluating non-deterministic LLM output: validators, regression fixtures from real failures, prompt testing.

LAFwords (lafwords.com): ten varied example sentences per word, generated by AI and checked by dictionaries and rules.

QA automation and AI quality roles. Talks about testing LLM-generated content. Honest feedback on LAFwords.