AI/LLM Project Evaluator at TURING (2026-12 – Present)
- Evaluate LLM responses using structured rubrics, with emphasis on accuracy, instruction-following, and safety/compliance.
- Review complex instruction-following notebooks and deliver clear, actionable feedback to improve output quality.
- Contribute to quality assurance and calibration by aligning on scoring standards, resolving ambiguous edge cases, and maintaining evaluator consistency.
- Document recurring failure patterns and propose guideline updates to improve evaluation reliability.
- Understand User prompt
- Task completion, factual accuracy, AI performance and overall justification of the template
- Designed/maintained evaluation rubrics and prompt suites covering real workflows (e.g., order flow, shipping logic, exceptions, and returns).
- Conducted error analysis (hallucinations, missing constraints, invalid assumptions) and summarized insights for stakeholders.
- Coordinated reviewer alignment through calibration sessions and quality checks to reduce scoring variance and raise consistency.
- Served as a reviewer of complex instruction-following notebooks, ensuring high standards for challenging evaluation cases.
Customer Success/Training Specialist at COURAGE POINT OF SALES (2024-10 – 2025-09)
- Strategic Account Enablement - Managed a portfolio of local merchants, conducting quarterly business reviews and training sessions to ensure feature adoption.
- Launch Support - Assisted in the Go-To-Market strategy for new features by creating local language manuals and How-To decks that enabled the Sales team to better communicate value propositions.
- Process Improvement - Implemented a new Learning Management System (LMS) for external users, tracking certification progress using Excel sheets and correlating trained merchants with a higher retention rate.
- Metrics & Reporting - Analyzed merchant engagement data to identify knowledge gaps, subsequently launching a "Tips & Tricks" newsletter with an open rate of 35%.
Support Officer at COURAGE POINT OF SALES (2021-12 – 2024-08)
- Front-line Problem Solving - Resolved an average of 17 merchant inquiries per week, gaining deep insight into common user pain points and roadblocks.
- Feedback Loop - Acted as the "Voice of the Merchant" in internal meetings, advocating for critical changes that simplified the checkout configuration process.
- Documentation - Initiated the team's internal resolution, documenting troubleshooting steps for complex integration issues
