Juan T.

Key Skills

Software

Scale AI

Remotasks

Top Subject Matter

Automotriz y Conducción Autónoma: Entrenamiento de modelos para detección de objetos, semáforos y segmentación de entornos urbanos.

Tecnología y Modelos de Lenguaje (LLM): Categorización de datos, evaluación de respuestas y entrenamiento de lógica para IAs conversacionales.

Logística y Transporte: Aplicación de conocimientos operativos de conducción para el etiquetado preciso de rutas y seguridad vial.

Top Data Types

Image

Video

3D Sensor

Top Task Types

Bounding Box

Segmentation

Cuboid

Polyline

Classification

Polygon

Text Generation

Object Detection

Prompt + Response Writing (SFT)

Freelancer Overview

Driver. Brings 18+ years of professional experience across complex professional workflows, research, and quality-focused execution. Education includes High School Diploma, U.E.M.R Rafael Taylhardat (2006) and Bachelor of Science, UNEFA (2007).

Labeling Experience

AI Training Specialist - LiDAR & 3D Point Cloud Annotation

VideoBounding Box

Execution of annotation and labeling tasks for 3D point clouds (LiDAR) for the training of autonomous driving algorithms. The work consisted of the precise identification and delimitation of objects (vehicles, pedestrians, infrastructure) using 3D cuboids, maintaining high standards of quality and spatial precision for road safety.

2020 - Present

Video Data Curator & Editor for Machine Learning

VideoObject Detection

Processing and editing video segments for computer vision datasets. Identifying specific actions in timelines and segmenting moving objects to improve the visual recognition of artificial intelligence models.

2026 - Present

LLM Training & Reinforcement Learning (RLHF)

TextClassification

Training large language models (LLMs) by categorizing responses, evaluating truthfulness, and ranking results based on utility and safety. Includes creative writing tasks and correction of logical errors to improve human-AI interaction.

2023 - 2023

Full Scene Semantic Segmentation for Urban Environments

VideoSegmentation

Detailed pixel-level labeling of complete urban scenes. The work included color-based segmentation of multiple object classes, including infrastructure (poles, signs), living beings (people, animals), and vehicles, allowing the AI to accurately identify the boundaries of each element on public roads.

2020 - 2023

Traffic Light Classification & State Identification

ImageClassification

Identification and classification of traffic lights in complex road environments. The task consisted of labeling the type of traffic light (vehicular, pedestrian, turning) and the precise state of the lighting (light color, directional arrows) to train real-time traffic signal detection models.

2018 - 2019

Education

Work History