Looking for a job

Eduarda Vilalba

@eduardavilalba

Message

LLM evaluator at Outlier and RWS Group, rating Brazilian Portuguese AI responses and writing training feedback.

Brazil

Message

What I'm looking for

I'm available for an immediate start.

At Outlier, I evaluated AI language model responses in Brazilian Portuguese for quality, helpfulness, accuracy, and naturalness. I wrote structured feedback and comparative rankings in English using detailed rubrics.

At RWS Group, I evaluated model responses across multiple task types and wrote feedback to support LLM training. I also audited transcription accuracy for Mercor / Amazon Sonic, checking speech against audio and verifying word-level timestamps.

Across projects with Micro1, CrowdGen / Appen, and iMerit Scholars, I collected egocentric video, assessed ad relevance, and annotated AI training data. I’m a native Brazilian Portuguese speaker with advanced English reading and writing skills, and experience with European Portuguese.

Experience

Work history, roles, and key accomplishments

Current

Transcription QA Auditor

Current

Mercor / Amazon Sonic

Oct 2026 - Present (0 months)

Audited transcriptions of human and agent speech in Brazilian Portuguese, applying error codes and pass/fail verdicts with written rationale. Performed forced alignment audits and corrected timestamp errors.

Transcription QA Error Code Application Forced Alignment Timestamp Correction Brazilian Portuguese English

Egocentric Video Data Collector

Micro1

Jun 2026 - Oct 2026 (4 months)

Recorded first-person egocentric video footage of real-world household tasks using a head-mounted device for AI training datasets.

Egocentric Video Recording Data Collection AI Training Data

LLM Evaluator

RWS Group

Jun 2026 - Oct 2026 (4 months)

Evaluated AI model responses in Brazilian Portuguese across multiple task types, applying quality guidelines and writing detailed feedback in English.

LLM Evaluation Brazilian Portuguese English Quality Guidelines

Ad Relevance Evaluator

CrowdGen / Appen

May 2026 - Oct 2026 (5 months)

Evaluated relevance and quality of social media ads for Meta, applying structured multi-criteria guidelines and providing written commentary.

Ad Relevance Assessment Content Evaluation Commentary Writing Social Media Ads

LLM Evaluator

Outlier

Apr 2026 - Oct 2026 (6 months)

Engaged with AI language models in real-time conversations in Brazilian Portuguese, evaluating response quality and writing structured feedback in English.

LLM Evaluation RLHF Brazilian Portuguese English Feedback Writing