Geoffrey O.

Key Skills

Software

Labelbox

Top Subject Matter

Computer vision -image annotation(bounding boxes,polygons,segmentations

Multimodal dataset labeling (image+text+audio)

Data preparation and labeling support

Top Data Types

Image

Text

Document

Top Task Types

Bounding Box

Segmentation

Classification

Freelancer Overview

AI Trainer – Data Labeling Focus (Payona, Kenya). Core strengths include Labelbox, N, and A. Education includes Bachelor of Science, University of Nairobi. AI-training focus includes data types such as Image, Text, and Document and labeling workflows including Bounding Box, Segmentation, and Classification.

Labeling Experience

AI Trainer – Data Labeling Focus (Payona, Kenya)

LabelboxImageBounding Box

Served as an AI Trainer focused on data labeling, performing computer vision annotations for client-ready datasets. Produced bounding boxes and polygons with high consistency while interpreting guidelines and applying feedback from reviews. Conducted secondary reviews on junior annotators to ensure compliance with labeling standards. • Annotate thousands of images per week using bounding boxes and polygons with >99% consistency • Perform secondary reviews to verify client guideline compliance • Create and update labeling playbooks to reduce ambiguity • Apply best practices to maintain quality metrics during ongoing annotation

2024 - Present

Freelance Data Labeler (Upwork / Local Platforms)

TextSegmentation

Completed freelance data labeling assignments for computer vision and NLP projects across multiple industries. Applied guideline interpretation to ensure labeled outputs met project requirements and client expectations. Iterated on labeling approaches to handle diverse data types and annotation specifications. • Deliver labeled datasets across e-commerce, healthcare, and autonomous systems • Adapt to 20+ labeling guideline sets with zero client rejections • Maintain consistent quality for speed, precision, and communication • Support both computer vision and NLP labeling tasks as required

2014 - 2020

Data Entry & Labeling Support

DocumentClassification

Provided data entry and labeling support by preparing raw datasets before annotation work began. Ensured dataset cleanliness by removing duplicates, standardizing file formats, and flagging edge cases for additional review. Tracked labeling progress and escalated difficult samples to the team to maintain overall annotation quality. • Remove duplicates and standardize raw data formats for labeling readiness • Flag edge cases for team review • Use Google Sheets to track labeling progress • Flag difficult samples for team escalation

2010 - 2014

Education

Work History