Regression testing is often tedious, time-consuming, and prone to bias. As PyCharm’s test plan grew to ~700 cases, manually covering everything before each release became unrealistic. The QA team faced a familiar dilemma: repeat what’s easy and known, or risk leaving blind spots untouched. Denis Mashutin will share how they tackled regression fatigue and routine bias with a simple but powerful idea — shuffled checks. By randomly sampling small chunks of their test plan for each release cycle, they improved coverage, found critical bugs earlier, and made testing less painful for the team. No fancy frameworks, no extra headcount — just practical automation and smarter processes anyone can adopt.