Quality Assurance Analyst

Lead Quality Assurance across multiple Safety and Model Labeling workflows, ensuring alignment with global policies and maintaining high performance standards across APAC markets. Conduct structured RCA deep-dives to identify systemic knowledge gaps and operational risks; develop targeted training interventions to improve accuracy and reduce repeat errors. Deliver weekly calibration and policy alignment sessions with moderators, TLs, and BPO teams; clarify complex policy interpretations and ensure consistent understanding across regions. Partner with Policy, Training, and Product Engineering teams to review guideline updates, translate policies into clear workflow documentation, and support safe implementation. Prepare data-driven insights using QA metrics, AHT analysis, and error trends to inform operational decisions and refine training strategies. Provide safety expertise on escalations involving high-risk or sensitive content, ensuring decisions meet global Safety Model standards.

Designed structured prompts for AI agents to improve moderation consistency and model accuracy. Collaborated closely with LLM and policy teams to align prompt logic with compliance and policy safety goals. Crafted effective prompts for US region AI models based on Model & Human PBRs to optimize model precision rates and improve LLM response quality. Developed clear, concise, and well-structured prompts tailored for various LLMs (e.g., GPT-3, GPT-4), incorporating techniques such as few-shot learning and chain-of-thought prompting. Analyzed AI model outputs to assess response quality, identify areas for improvement, and iteratively refine prompt structures. Designed prompt engineering workflows to meet 90% precision targets, aligning with PBR requirements and performance KPIs. Collaborated with cross-functional teams to integrate prompt solutions into development pipelines, utilizing Python and JSON for testing and validation.

Oversaw BPO project completion and resource planning, ensuring sufficient headcount to support all MLM & LLM projects. Monitored BPO project metrics (accuracy, AHT, UR, occupancy) and implemented action plans to address low performance or high AHT. Delivered SOP and PBR training for new project intakes, enabling BPO partners to achieve ≥ 99% accuracy on first deliverables. Provided continuous support and calibration sessions with BPO teams to discuss errors and edge cases, reducing rework by 30%. Set market-specific AHT targets through task simulations and performance modeling to streamline operations. Served as backup Market POC for UK, PH, and SG markets during IHQAs’ absence, ensuring seamless communications between stakeholders. Collaborated with project owners and policy teams to conduct detailed RCAs, raise PBR/SOP concerns, and resolve challenges, maintaining high project quality across markets.

Organized and detail-oriented with strong work ethic. Learned and adapted quickly to new technology and software applications. Provided professional services and support in a dynamic work environment. Participated in team projects, demonstrating the ability to work collaboratively and effectively. Acted as team leader in group projects, delegating tasks and providing feedback. Handover the project to US POCs (conduct a session to convey the tasks).

Delivered onboarding and ongoing training for new and experienced moderators on Safety policies, enforcement workflows, and operational procedures. Created high-quality training decks, policy guides, RCA documentation, and standard workflows to support consistent knowledge uptake. Addressed complex policy escalations, conducted deep-dives on high-risk content trends, and collaborated with Policy teams to improve guideline clarity. Coached moderators through 1:1 sessions, provided structured feedback, and contributed to overall quality uplift across regional teams. Improved operational accuracy through SOP updates, policy clarification, and streamlined escalation protocols. Created multitagging labels and compiled calibrated cases for better visibility.

Reviewed high-risk and sensitive content, including hate speech, harassment, child safety, bullying, violent acts, and self-harm across diverse markets. Ensured accurate and consistent enforcement of community guidelines while maintaining reviewer wellbeing and operational integrity. Supported early-stage training improvements by providing policy feedback, real-time insight sharing, and trend documentation for safety teams. Handled sensitive edge cases and collaborated with cross-functional teams to ensure safe and compliant decision-making.