Pedro Ruiz

Key Skills

Software

AWS SageMaker

Labelbox

SuperAnnotate

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Classification

Object Detection

Segmentation

Freelancer Overview

I am a (English/Spanish) Data Labeling Specialist with hands-on experience in annotating image, video, and text datasets for AI training. Since 2023, I’ve worked remotely with early-stage startups on projects involving product recognition in retail, human action and emotion detection in video, and natural language processing tasks like sentiment analysis and entity recognition. I’ve used tools such as Labelbox, SuperAnnotate, and Amazon SageMaker Ground Truth, and consistently met high-quality standards under tight deadlines. My strength lies in combining technical precision with language skills and a strong understanding of annotation guidelines. I adapt quickly to different data types and labeling frameworks, and I’ve contributed both as an annotator and as part of QA review processes. I’m passionate about supporting AI development through clean, accurate, and consistent training data.

Labeling Experience

Freelance NLP Data Labeler

LabelboxTextSegmentationClassification

I collaborated on a small-scale text annotation task aimed at training a language model for HR-related applications. The project included labeling approximately 1,500 text samples in English and Spanish using Labelbox. Tasks involved classifying sentiment, identifying user intent, and tagging named entities within short text segments like emails and job ads. The annotation team worked with clear but evolving guidelines, and quality was monitored through random review batches with a 90–95% consistency goal.

I collaborated on a small-scale text annotation task aimed at training a language model for HR-related applications. The project included labeling approximately 1,500 text samples in English and Spanish using Labelbox. Tasks involved classifying sentiment, identifying user intent, and tagging named entities within short text segments like emails and job ads. The annotation team worked with clear but evolving guidelines, and quality was monitored through random review batches with a 90–95% consistency goal.

2024 - 2024

Freelance Data Annotator

SuperannotateImageSegmentationClassification

This project involved annotating images and short video clips of retail shelves for a prototype product recognition system. The scope included labeling a few hundred product instances across different categories using SuperAnnotate. I performed bounding box and polygon segmentation to classify items by brand and type, and helped track products across video frames. The dataset was used to test early-stage object detection models. Quality checks were done weekly, with an expected accuracy threshold of 95%, and revisions were provided for corrections or unclear cases.

This project involved annotating images and short video clips of retail shelves for a prototype product recognition system. The scope included labeling a few hundred product instances across different categories using SuperAnnotate. I performed bounding box and polygon segmentation to classify items by brand and type, and helped track products across video frames. The dataset was used to test early-stage object detection models. Quality checks were done weekly, with an expected accuracy threshold of 95%, and revisions were provided for corrections or unclear cases.

2024 - 2024

Education

Work History