Margaret Murunga

Key Skills

Software

Labelbox

CVAT

Prodigy

Clickworker

CloudFactory

Data Annotation Tech

Dataloop

Img Lab

Label Studio

LabelImg

LightTag

Mercor

Mindrift

Top Subject Matter

Natural Language Processing

Chatbots Domain Expertise

Generative AI

Top Data Types

Text

Image

Top Task Types

Entity Ner Classification

Bounding Box

Freelancer Overview

AI Data Labeler & Trainer (Freelance / Remote). Brings 18+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Labelbox, CVAT, and Prodigy. Education includes Doctor of Philosophy, Stanford University (2009) and Master of Science, University of Illinois Urbana-Champaign (2006). AI-training focus includes data types such as Text and Image and labeling workflows including Entity (NER) Classification, Bounding Box, and Evaluation.

Labeling Experience

AI Data Labeler & Trainer (Freelance / Remote)

CVATImageQuestion Answering

Labeled images as part of AI data annotation projects, performing tasks such as image tagging and object detection. Delivered high-quality bounding box and segmentation annotations according to established guidelines. Contributed to refining team annotation practices and tool workflows. • Produced consistent, high-precision bounding boxes and segmented images. • Ensured guidelines were adhered to throughout project duration. • Leveraged Labelbox and CVAT for efficient image annotation workflows. • Supported datasets used in computer vision and generative AI models.

Labeled images as part of AI data annotation projects, performing tasks such as image tagging and object detection. Delivered high-quality bounding box and segmentation annotations according to established guidelines. Contributed to refining team annotation practices and tool workflows. • Produced consistent, high-precision bounding boxes and segmented images. • Ensured guidelines were adhered to throughout project duration. • Leveraged Labelbox and CVAT for efficient image annotation workflows. • Supported datasets used in computer vision and generative AI models.

2024 - Present

AI Data Labeler & Trainer (Freelance / Remote)

LabelboxTextPoint Key Point

As a freelance AI Data Labeler & Trainer, I annotated and labeled large-scale text datasets for NLP and computer vision projects. I reviewed AI-generated text outputs and applied consistent labeling guidelines to ensure data quality. I contributed to training datasets for chatbot and generative AI systems across more than 10,000 entries. • Tasks included text classification, sentiment labeling, and named entity recognition (NER). • Conducted quality checks and resolved annotation inconsistencies. • Utilized multiple software tools including Labelbox, CVAT, Prodigy, and Scale AI. • Ensured high accuracy, consistency, and bias reduction in data preparation.

As a freelance AI Data Labeler & Trainer, I annotated and labeled large-scale text datasets for NLP and computer vision projects. I reviewed AI-generated text outputs and applied consistent labeling guidelines to ensure data quality. I contributed to training datasets for chatbot and generative AI systems across more than 10,000 entries. • Tasks included text classification, sentiment labeling, and named entity recognition (NER). • Conducted quality checks and resolved annotation inconsistencies. • Utilized multiple software tools including Labelbox, CVAT, Prodigy, and Scale AI. • Ensured high accuracy, consistency, and bias reduction in data preparation.

2022 - Present

AI Data Labeler & Trainer (Freelance / Remote)

ProdigyTextFine Tuning

I performed evaluation of large language model outputs, assessing responses for correctness, bias, and overall coherence. This involved reviewing generated text against detailed quality and ethical standards. My work ensured improved reliability and safety in AI model deployments. • Rated and evaluated LLM and chatbot outputs for various user prompts. • Provided feedback on output quality and guideline adherence. • Helped identify and reduce sources of AI-generated bias. • Supported AI training pipelines with robust evaluation data.

I performed evaluation of large language model outputs, assessing responses for correctness, bias, and overall coherence. This involved reviewing generated text against detailed quality and ethical standards. My work ensured improved reliability and safety in AI model deployments. • Rated and evaluated LLM and chatbot outputs for various user prompts. • Provided feedback on output quality and guideline adherence. • Helped identify and reduce sources of AI-generated bias. • Supported AI training pipelines with robust evaluation data.

2014 - Present

Education

Work History