Thipsuda Thaengthong

AI Data Evaluator & Prompt Engineer (Contract) at DataAnnotation (2024-06 – 2026-07)

Evaluated AI responses in multi-turn conversations, focusing on memory retention, contextual understanding, personalization, and natural Thai language quality.
Performed A/B testing to compare model outputs, assessing logical reasoning, instruction following, accuracy, and overall response quality.
Designed creative adversarial prompts specifically tailored to stress-test model limitations in semantic analysis, contextual localization, and natural language generation, ensuring AI outputs maintain cultural accuracy, tone, and appropriate Thai idiomatic expression.
Documented evaluation findings, explained model errors, and provided improved reference responses to support model training.
Developed detailed evaluation rubrics and clear scoring criteria to improve consistency across annotation and review tasks.
Reviewed teammates' work, provided feedback, and helped maintain high-quality annotation standards.

Search Engine Evaluator (Contract) at Appen (2018-03 – 2024-03)

Evaluated search engine results based on relevance, user intent, and overall search quality.
Assessed the quality and relevance of online advertisements across major social media platforms.
Verified information from multiple online sources to evaluate credibility and identify misleading or low-quality content.

Freelance Translator & Content Writer (English ↔ Thai) (2012 – Present)

Translate and localize English and Thai content for online communities and social media, including real-time news updates and live event coverage.
Adapt complex information into natural, easy-to-understand Thai while preserving meaning, tone, and cultural context.
Verify information across multiple sources before publication to ensure accuracy and consistency.

เกี่ยวกับ