Davianys Zapata

Key Skills

Software

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Image

Text

Top Task Types

Computer Programming Coding

Prompt Response Writing SFT

Red Teaming

Freelancer Overview

I have hands-on experience in AI training and model evaluation through my work with AI companies, where I evaluated large language models by stress-testing their reasoning, safety, and robustness. This included intentionally inducing failure cases, identifying hallucinations, bias, and unsafe behaviors, and conducting red-teaming exercises to assess how models respond to adversarial prompts and edge cases. I provided structured, high-quality feedback to improve model performance, alignment, and safety across a wide range of scenarios in both English and Spanish. My background in cybersecurity, chemistry, and data science sets me apart by bringing a strong analytical and adversarial mindset to AI training tasks. I am skilled at breaking down complex outputs, validating factual and logical consistency, and applying security and scientific rigor to annotation and evaluation workflows.

Labeling Experience

Code Red Teaming

Scale AIComputer Code ProgrammingRed Teaming

The goal of this project was to red team team a model to generate harmful code.

The goal of this project was to red team team a model to generate harmful code.

2025 - 2025

Data Science project

Scale AIDocumentPrompt Response Writing SFT

The task was to create prompts that require a model to input and process one or multiple different files and produce one or multiple functional file outputs.

The task was to create prompts that require a model to input and process one or multiple different files and produce one or multiple functional file outputs.

2024 - 2025

Education

Work History