For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
Catarina Guimarães

Catarina Guimarães

LLM Evaluation, RLHF, Audio

PORTUGAL flag
Braga, Portugal
$20.00/hrIntermediateAppenLabelboxScale AI

Key Skills

Software

AppenAppen
LabelboxLabelbox
Scale AIScale AI

Top Subject Matter

No subject matter listed

Top Data Types

AudioAudio
ImageImage
TextText

Top Label Types

Audio Recording
Classification
Evaluation Rating
Prompt Response Writing SFT
RLHF

Freelancer Overview

I have recent experience in AI Training Data and labeling. This experience encompasses several key areas of Language Model (LLM) development and safety. A significant qualification is my involvement in Reinforcement Learning from Human Feedback (RLHF), serving as both an attempt-maker and a reviewer. My skills are further differentiated by my involvement in multi-modal and localization tasks. This includes voice acting recording and reviewing and audio and video annotation. With Appen, I am specialized in translations between Portuguese (pt) and English (en) , lip sync evaluation , and labeling safety prompts/responses. This extensive and varied freelance experience highlights my adaptability and commitment to various critical aspects of data preparation and model refinement.

IntermediateEnglishPortuguese

Labeling Experience

Appen

Video and audio annotation and Lip sync evaluation

AppenVideoTranslation Localization
The project involed to annotate the accuracy of the translation. The prosody, voice similarity and naturalness of the translated video and if it falls under the project guidelines.

The project involed to annotate the accuracy of the translation. The prosody, voice similarity and naturalness of the translated video and if it falls under the project guidelines.

2025
Scale AI

Scale AI

Scale AIImagePrompt Response Writing SFT
In this project we had to choose a image from the project image archive respecting the project usecase. Than a prompt was written and a response, after we had to evaluate the response written and the one created by the model. The response we wrote must be better than the one the model wrote.

In this project we had to choose a image from the project image archive respecting the project usecase. Than a prompt was written and a response, after we had to evaluate the response written and the one created by the model. The response we wrote must be better than the one the model wrote.

2025 - 2025

Education

U

University of Minho

Honours Bachelor's Degree in Education, Education

Honours Bachelor's Degree in Education
2005 - 2010

Work History

J

Janes Concept Store

Sales Assistant

Braga
2018 - Present
L

Lugar da Jóia

Sales Assistant

Braga
2016 - 2018