Maria Jose Gonzalez Fonseca

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Text

Video

Top Task Types

Audio Recording

Evaluation Rating

Prompt Response Writing SFT

Question Answering

RLHF

Freelancer Overview

I'm an AI Quality Auditor and AI Trainer with solid experience evaluating, improving, and training large language models through detailed review processes. I specialize in analyzing AI-generated responses, identifying errors, assessing accuracy, and ensuring compliance with project standards. My work includes reviewing complex tasks, detecting inconsistencies, and guiding models toward safer, clearer, and more aligned outputs. I have also contributed to the creation, refinement, and application of rubrics used to evaluate model performance. This includes setting quality criteria, defining scoring guidelines, and ensuring consistent evaluations across diverse datasets. My experience working with proprietary annotation platforms has strengthened my ability to deliver precise, well-structured, and high-impact feedback essential for developing reliable AI systems.

Labeling Experience

Rubric Creation and Evaluation Framework Design

Scale AITextRLHFEvaluation Rating

Developed and refined rubrics used to evaluate model responses. Defined scoring criteria, quality levels, and evaluation rules to ensure consistency and clarity. Reviewed pilot datasets to validate rubric effectiveness, identified ambiguous areas, and recommended improvements. Applied the final rubric at scale to maintain accurate and standardized evaluations across thousands of samples.

Developed and refined rubrics used to evaluate model responses. Defined scoring criteria, quality levels, and evaluation rules to ensure consistency and clarity. Reviewed pilot datasets to validate rubric effectiveness, identified ambiguous areas, and recommended improvements. Applied the final rubric at scale to maintain accurate and standardized evaluations across thousands of samples.

2025 - 2025

AI Training Support and Instruction Testing

Scale AITextRLHFFine Tuning

Conducted supervised fine-tuning tasks by reviewing instructions, testing model follow-through, and providing structured feedback. Evaluated whether the model accurately followed task requirements and aligned with desired outcomes. Identified inconsistencies and improvement opportunities to strengthen instruction-following behavior. Contributed to training cycles that improved model alignment and reliability.

Conducted supervised fine-tuning tasks by reviewing instructions, testing model follow-through, and providing structured feedback. Evaluated whether the model accurately followed task requirements and aligned with desired outcomes. Identified inconsistencies and improvement opportunities to strengthen instruction-following behavior. Contributed to training cycles that improved model alignment and reliability.

2024 - 2025

AI Quality Auditing & Response Evaluation

Scale AITextRLHFEvaluation Rating

Reviewed AI-generated responses across multiple tasks, checking for accuracy, clarity, safety, and consistency. Identified errors, flagged critical issues, and ensured that outputs met strict quality standards. Provided corrective guidance to improve model reasoning and performance. Delivered high-volume evaluations using detailed instructions and strict quality guidelines.

Reviewed AI-generated responses across multiple tasks, checking for accuracy, clarity, safety, and consistency. Identified errors, flagged critical issues, and ensured that outputs met strict quality standards. Provided corrective guidance to improve model reasoning and performance. Delivered high-volume evaluations using detailed instructions and strict quality guidelines.

2024 - 2025

Education

Work History