Tarso Leirias
AI & Language Specialist | Portuguese & English | AI Evaluation | Multimodal Content
Talento
Porto AlegreMembro desde 20 de outubro de 2026
AI TrainingCustomer Relations
·
Contratar esta pessoa
Envie uma proposta de trabalho diretamente para este candidato
AI TrainingCustomer Relations
Sobre
Detail-oriented professional with extensive background in customer relationship managment, insurance product development and general data analysis. Currently looking for remote work opportunities. Recent work experience in AI Training Data Annotation for LLMs in Brazillian Portuguese and American English, and looking to further my experience in this field.
Experiência
Operations Manager at Garantia Insurance Brokers (2022-03 – Present)
- Streamlined personalized customer support services, increasing customer satisfaction and attention to compliance;
- Consistently fast managing of customer demand;
- Empower customers to resolve frequent issues independently through searchable guides.
AI Model Training Annotator at Aether Project, through Outlier (2026-08 – Present)
- Contributor for AI Model Training tasks, AI-generated content evaluation, multilingual projects, multimodal LLms, and.
- Evaluate and compare AI-generated outputs according to specific quality criteria, rubrics, and task requirements.
- Applied skills in native pt-BR and advanced en-US comprehension, including grammar, writing, reading analysis, and communication.
- Evaluate Brazilian Portuguese content, including grammar, vocabulary, spelling, syntax, terminology, idiomatic expressions, formal and informal communication, and regional variations.
- Evaluate output from multimodal AI agents using project-specific criteria across multiple LLM comparison projects. Maintaining accuracy, consistency, attention to detail, and adherence to complex evaluation guidelines in remote project based environments.
Formação
Foundation in Computer Sciences – Lutheran University of Brasil - ULBRA (2010-03 – 2012-10)
Competências
AI & Language SpecialistPortuguese & EnglishAI EvaluationMultimodal Content





