Gabriela I.

Key Skills

Software

Other

Top Subject Matter

AI response evaluation

Annotation Domain Expertise

and language quality review (Romanian-English)

Top Data Types

Text

Top Task Types

Bounding Box

Object Detection

Question Answering

Transcription

Data Collection

Freelancer Overview

Freelance AI Data Annotator / AI Trainer (Outlier) — Remote. Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Higher National Certificate, The University of the West of Scotland (2026) and Level 5 Certificate in Dental Nursing, New College Lanarkshire (2025). AI-training focus includes data types such as Text and labeling workflows including Evaluation and Rating.

Labeling Experience

Freelance AI Data Annotator / AI Trainer (Outlier) — Remote

OtherText

Freelance AI Data Annotator/AI Trainer experience evaluating AI-generated responses against task-specific guidelines, rubrics, and quality criteria. Responsibilities included comparing outputs for prompt adherence, user intent, safety, tone, completeness, and relevance while checking grammar, clarity, and factual consistency. Work also involved identifying unsupported claims and low-quality responses, and applying Romanian-English judgment for meaning and nuance. • Evaluated prompt-response quality using rubrics, guidelines, and edge-case awareness • Performed instruction-following QA, hallucination spotting, and relevance/completeness assessments • Provided quality checks for tone, cultural nuance, and natural phrasing in Romanian-English • Completed AI annotation-style tasks to improve model quality and consistency

2023 - Present

Freelance AI Data Annotator / AI Trainer - Outlier (Remote, Freelance)

OtherText

Freelance AI data annotator and AI trainer work for Outlier focused on evaluating and ranking AI-generated responses against task-specific guidelines and rubrics. The work included quality checks for prompt adherence, instruction-following, relevance, tone, and factual consistency, with identification of unsupported claims and low-quality outputs. Romanian-English language judgment was applied to assess meaning accuracy, naturalness, and nuance while documenting findings for labeling tasks. • Evaluated prompt-response quality using rubrics and quality criteria • Compared responses for adherence, intent, safety, completeness, and relevance • Reviewed grammar, clarity, and factual consistency; flagged hallucinations/unsupported claims • Completed annotation-style labeling tasks to support AI model improvement

2023 - Present

Education

Work History