LLM Evaluation Specialist

I have evaluated more than 1,000 outputs for a confidential frontier AI lab program across increasingly complex evaluation phases. My work began with selecting best responses and identifying grammar and spelling issues in real user-model conversations.

I conduct multilingual search evaluation by authoring themed prompts, testing them across four models, and scoring source reliability, localization, and time correctness. I also design flawed-response prompts, step-by-step guided instructions, and rubrics for self-correction testing, while evaluating AI speech for naturalness, intelligibility, pronunciation, transcript alignment, and linguistic quality in Spanish and English.

I transcribed and quality-checked AI-generated audio and speech content against defined accuracy and standardization criteria.

I identified transcription errors and content-quality issues and delivered structured feedback to improve output consistency and quality.

I performed quality-control testing, sampling, and structured data analysis against standardized specifications in an industrial agricultural-processing environment.

I investigated quality deviations, documented findings, and supported corrective actions through root-cause analysis. This experience developed rigorous QA, error-detection, and assessment practices transferable to AI evaluation.

Completed Agricultural Engineering degree. Developed scientific and analytical training in structured problem-solving, technical assessment, and agricultural-domain evaluation.