I manually tested an AI meeting assistant by checking expected versus actual behavior across user workflows. I captured screenshots and reproducible evidence to document deviations and support issue review.
At GlobalLogic, I evaluated AI and data outputs against detailed acceptance criteria, maintaining approximately 98% quality accuracy while handling 180+ evaluation events per day. I documented edge cases and recurring failure patterns, and escalated ambiguous outcomes with evidence.
At Foundever, I performed structured QA evaluations and tracked recurring issues against defined standards. I also coached agents and trained new hires using documented quality findings.
