Bruno Joshua

Key Skills

Software

CVAT

Labelbox

Mindrift

Top Subject Matter

No subject matter listed

Top Data Types

Text

Video

Top Label Types

Evaluation Rating

Classification

RLHF

Freelancer Overview

I am an AI data annotator and trainer with extensive experience in NLP, search quality evaluation and content moderation. Over the past several years, I’ve worked on projects that involve large-scale data labeling, intent classification, sentiment analysis and computer vision annotation. I specialize in improving AI models for search engines, chatbots and social media platforms by ensuring data accuracy, contextual understanding and safety compliance. I am fluent in English, Hindi, and Swahili, and thrive in fast-paced, fully remote environments where data integrity and precision are critical.

Labeling Experience

AI Data Annotator

LabelboxImageBounding BoxPolygon

I annotated and labeled 100+ hours of image data using Labelbox to support AI and NLP model training. The work included bounding box, polygon, segmentation, object detection and classification tasks, as well as RLHF-based evaluation and quality checks.I ensured high accuracy and consistency by strictly following labeling guidelines, reviewing annotations for correctness and addressing edge cases. Collaboration with AI engineers and researchers allowed the dataset to directly improve model performance, contextual understanding and decision-making for computer vision and LLM applications.

I annotated and labeled 100+ hours of image data using Labelbox to support AI and NLP model training. The work included bounding box, polygon, segmentation, object detection and classification tasks, as well as RLHF-based evaluation and quality checks.I ensured high accuracy and consistency by strictly following labeling guidelines, reviewing annotations for correctness and addressing edge cases. Collaboration with AI engineers and researchers allowed the dataset to directly improve model performance, contextual understanding and decision-making for computer vision and LLM applications.

2025 - 2025

Human Feedback-Based LLM Training

CVATVideoClassificationRLHF

orked on a short-term, high-volume AI training project focused on improving large language models using human feedback. The project involved annotating 100 hours of car video data by classifying objects, scenes, and human interactions within video frames.

orked on a short-term, high-volume AI training project focused on improving large language models using human feedback. The project involved annotating 100 hours of car video data by classifying objects, scenes, and human interactions within video frames.

2025 - 2025

AI Trainer & Content Moderator

MindriftTextEvaluation Rating

Evaluated AI-generated responses and search results, ensuring high-quality, accurate, and contextually appropriate content.

Evaluated AI-generated responses and search results, ensuring high-quality, accurate, and contextually appropriate content.

2018 - 2020

Education

Work History