Expectation
Establish the expected behaviour from requirements, designs or policies. Clarify gaps with the team, and investigate automated test results before deciding what they show.
Pelin Dragos
Quality assurance · GRC
QA experience · 5+ years · EA Games, Funcom and TikTok
Clear test cases. Reproducible defects. Useful reports.
Functional, regression and exploratory testing, with clear test cases, reproducible defect reports and analysis of automated test results. My experience includes FIFA testing at EA Games and complex game features at Funcom.
How I approach testing
From expected behaviour to verification
Reproduction
Environment, build, data and the steps, written so the next person sees the same thing instead of a conclusion in a screenshot. Intermittent issues should include the available evidence and the conditions observed, even when they cannot be reproduced reliably.
Closure
Explain the impact, agree the severity and record what needs to change. Retest the fix and check related behaviour before closing the issue.
The same discipline has a public instance on this site: the build gates described under How this site is made. A criterion is written down, a script checks it, and the build fails when the two disagree.
The record
EA Games FIFA testing; Funcom feature work is covered under Gaming QA
EA Games
- Developed and executed manual black-box test scenarios for FIFA and analysed the results of automated tests.
- Created and maintained test cases in TestRail.
- Documented and tracked defects in Jira.
Content and data quality at TikTok
Applying written policies consistently
At TikTok, I review content and data quality against internal policies, assess vendor outputs and lead calibration meetings. Reporting and feedback help teams interpret requirements consistently and address recurring issues.
TikTok
- Review platform content and data against internal policies and quality standards, supporting the quality of AI training datasets.
- Review and correct moderation outputs from third-party vendors, checking adherence to internal requirements and safety guidelines.
- Lead calibration meetings with internal teams and external partners to align policy interpretation and address inconsistent decisions.
- Prepare performance reports and key performance indicators to monitor vendor quality and identify recurring data-labelling issues.
- Coordinate feedback with stakeholders and vendors to clarify quality requirements and support process improvement.
Governance, risk and compliance is where this method goes next. What the GRC study covered.