Greig Okombe

Key Skills

Software

Labelbox

Other

Top Subject Matter

Computer Vision

Medical Imaging

Retail Product Recognition

Top Data Types

Image

Text

Audio

Top Task Types

Bounding Box

Entity (NER) Classification

Classification

Polygon

Transcription

Freelancer Overview

Data Annotation Intern. Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Labelbox, Other, and VGG Image Annotator (VIA). Education includes Bachelor of Science, University of Nairobi (2024). AI-training focus includes data types such as Image, Text, and Medical and labeling workflows including Bounding Box, Entity (NER) Classification, and Classification.

Labeling Experience

Swahili Sentiment Analysis Dataset (Final Year Project)

OtherTextClassification

Constructed a sentiment analysis dataset with over 12,000 Swahili tweets for use in AI training. Designed sentiment labels (positive, negative, neutral) according to a fixed schema and led a team of annotators in executing the task. Ensured high quality through multi-stage annotation checks and documented procedures.• Established clear annotation guidelines with edge-case resolution• Oversaw annotation consensus processes among team members• Ensured dataset package included codebook, guidelines, and quality metrics• Applied Python tools for cleaning and data exportation

Constructed a sentiment analysis dataset with over 12,000 Swahili tweets for use in AI training. Designed sentiment labels (positive, negative, neutral) according to a fixed schema and led a team of annotators in executing the task. Ensured high quality through multi-stage annotation checks and documented procedures.• Established clear annotation guidelines with edge-case resolution• Oversaw annotation consensus processes among team members• Ensured dataset package included codebook, guidelines, and quality metrics• Applied Python tools for cleaning and data exportation

2024 - 2024

Audio Annotation (Speech-to-Text and Speaker Diarisation)

OtherAudioTranscription

Performed speech-to-text transcription and speaker diarisation for Swahili language datasets to support automatic speech recognition AI projects. Consistently tagged segments with emotion, tone, and noise level labels across multiple hours of audio recordings. Participated in routine quality checks to uphold strict annotation standards.• Annotated multiple hours of audio with speaker and emotion tags• Supported development of emotion recognition systems• Ensured annotation quality through inter-rater reliability reviews• Helped maintain guideline clarity and annotation consistency

Performed speech-to-text transcription and speaker diarisation for Swahili language datasets to support automatic speech recognition AI projects. Consistently tagged segments with emotion, tone, and noise level labels across multiple hours of audio recordings. Participated in routine quality checks to uphold strict annotation standards.• Annotated multiple hours of audio with speaker and emotion tags• Supported development of emotion recognition systems• Ensured annotation quality through inter-rater reliability reviews• Helped maintain guideline clarity and annotation consistency

2023 - 2023

Medical Image Annotation (Personal Project)

Polygon

Annotated chest X-ray and MRI images for pneumonia region detection as a self-directed project. Used polygon masks and bounding boxes to label regions of interest with high precision for use in AI model training. Managed a dataset of over 2,000 images and exported annotations in COCO JSON format.• Used VGG Image Annotator (VIA) for precise polygon annotation• Wrote reproducible annotation guidelines and instructions• Coordinated with 3 other annotators for consistency and quality• Focused on medical imaging challenges and region-level accuracy

Annotated chest X-ray and MRI images for pneumonia region detection as a self-directed project. Used polygon masks and bounding boxes to label regions of interest with high precision for use in AI model training. Managed a dataset of over 2,000 images and exported annotations in COCO JSON format.• Used VGG Image Annotator (VIA) for precise polygon annotation• Wrote reproducible annotation guidelines and instructions• Coordinated with 3 other annotators for consistency and quality• Focused on medical imaging challenges and region-level accuracy

2023 - 2023

Data Annotation Intern

LabelboxTextEntity Ner Classification

Labeled multilingual text datasets for intent classification, sentiment analysis, and named entity recognition. Applied detailed annotation guidelines to ensure high consistency in sentiment and intent tagging across both English and Swahili language corpora. Supported the development and training of NLP models for chatbot and conversational AI use cases.• Used multiple rounds of review to ensure label accuracy and agreement• Collaborated with peers on ambiguous segments and guideline updates• Performed entity recognition and sentiment tagging for social media and conversation data• Contributed to guideline improvements based on annotation feedback

Labeled multilingual text datasets for intent classification, sentiment analysis, and named entity recognition. Applied detailed annotation guidelines to ensure high consistency in sentiment and intent tagging across both English and Swahili language corpora. Supported the development and training of NLP models for chatbot and conversational AI use cases.• Used multiple rounds of review to ensure label accuracy and agreement• Collaborated with peers on ambiguous segments and guideline updates• Performed entity recognition and sentiment tagging for social media and conversation data• Contributed to guideline improvements based on annotation feedback

2023 - 2023

Data Annotation Intern

LabelboxImageBounding Box

Labeled over 50,000 images for object detection, segmentation, and pose estimation tasks supporting computer vision AI systems. Used bounding boxes, polygons, polylines, and semantic masks across diverse image categories including street scenes, retail, and medical scans. Utilized industry-standard annotation platforms to maintain rigorous data quality and throughput across all projects.• Maintained 98%+ annotation accuracy rate• Performed peer reviews and conducted inter-annotator agreement checks• Applied semantic and instance segmentation with quality assurance• Ensured annotations met evolving guideline specifications

Labeled over 50,000 images for object detection, segmentation, and pose estimation tasks supporting computer vision AI systems. Used bounding boxes, polygons, polylines, and semantic masks across diverse image categories including street scenes, retail, and medical scans. Utilized industry-standard annotation platforms to maintain rigorous data quality and throughput across all projects.• Maintained 98%+ annotation accuracy rate• Performed peer reviews and conducted inter-annotator agreement checks• Applied semantic and instance segmentation with quality assurance• Ensured annotations met evolving guideline specifications

2023 - 2023

Education

Work History