I am an AI evaluator, data trainer, scientific educator, and applied researcher based in Pelotas, Brazil. I specialize in rubric-based LLM evaluation, multilingual quality analysis, structured feedback, factuality review, instruction-following assessment, and human-in-the-loop quality workflows.
I currently work remotely on AI evaluation and data-training assignments for FlyRank AI and SME Solutions Group / Blue Mantis. My work involves reviewing AI-generated responses and textual data for clarity, reasoning, relevance, consistency, completeness, accuracy, and compliance with project guidelines.
I have also completed project-based AI evaluation work for Mercor and Alignerr, including model-output comparison, evidence-based response ranking, and audio-related aviation communication evaluation. I am comfortable documenting quality criteria, identifying unsupported claims, and producing concise, actionable feedback.
My academic background is in Chemistry and Science and Technology in Education. I have completed funded research projects involving Moodle learning objects, digital tools for science education, active methodologies, chemistry simulators, and AI-supported teacher education.
As an independent educator and academic reviewer, I have reviewed more than 140 undergraduate theses and trained 71 educators in AI literacy. I create educational materials, workshops, instructional sequences, and research-based content with attention to clarity, accessibility, responsible AI use, and evidence quality.
I am fluent in Portuguese, English, and German, and I am available for international remote work, contract projects, and multidisciplinary collaborations involving AI quality, scientific content, education, data training, and technical research.