Benard Odunga

Key Skills

Software

Sama

Appen

CVAT

Top Subject Matter

Autonomous Vehicles

Natural Language Processing

Government & NGO Data

Top Data Types

Image

Text

Document

Top Task Types

Bounding Box

Entity Ner Classification

Classification

Freelancer Overview

Senior Data Annotation Specialist. Core strengths include Sama, Appen, and CVAT. Education includes Bachelor of Arts, University of Nairobi (2017). AI-training focus includes data types such as Image and Text and labeling workflows including Bounding Box, Entity (NER) Classification, and Classification.

Labeling Experience

Senior Data Annotation Specialist

SamaImageBounding Box

Oversaw and contributed to bounding box, polygon, and semantic segmentation for large-scale computer vision datasets in the autonomous vehicle domain. Maintained annotation accuracy above 98% and produced over 450,000 high-quality labeled assets using Sama, Labelbox, and CVAT. Led the team on reinforcement learning from human feedback (RLHF) tasks and conducted extensive QA using internal and external audit tools.• Led team of 12 annotators on high-volume projects • Contributed to RLHF for large language models • Maintained 98.7% accuracy score • Developed annotation and style guidelines

Oversaw and contributed to bounding box, polygon, and semantic segmentation for large-scale computer vision datasets in the autonomous vehicle domain. Maintained annotation accuracy above 98% and produced over 450,000 high-quality labeled assets using Sama, Labelbox, and CVAT. Led the team on reinforcement learning from human feedback (RLHF) tasks and conducted extensive QA using internal and external audit tools.• Led team of 12 annotators on high-volume projects • Contributed to RLHF for large language models • Maintained 98.7% accuracy score • Developed annotation and style guidelines

2022 - Present

Data Annotation Specialist

AppenTextEntity Ner Classification

Executed text annotation tasks including NER, sentiment analysis, intent classification, and coreference resolution for NLP model training using Appen and related platforms. Provided transcription and speaker diarization for multilingual audio datasets, supporting the creation of robust speech and NLP systems. Flagged and reported data quality issues and beta-tested annotation tools for platform improvement.• Annotated for NER and sentiment analysis • Labeled audio for speaker diarization • Contributed feedback to tool development • Flagged >1,200 data quality issues

Executed text annotation tasks including NER, sentiment analysis, intent classification, and coreference resolution for NLP model training using Appen and related platforms. Provided transcription and speaker diarization for multilingual audio datasets, supporting the creation of robust speech and NLP systems. Flagged and reported data quality issues and beta-tested annotation tools for platform improvement.• Annotated for NER and sentiment analysis • Labeled audio for speaker diarization • Contributed feedback to tool development • Flagged >1,200 data quality issues

2020 - 2022

Data Entry & Annotation Associate

CVATImageClassification

Conducted data entry, document digitization, and annotated images and documents using CVAT and Label Studio for government and NGO projects. Achieved a 97% task acceptance rate and completed more than 40,000 micro-tasks during a six-month period. Developed early proficiencies in industry-standard annotation tools to meet client and project requirements.• Tagged and digitized images/documents • Used CVAT and Label Studio platforms • Maintained high task acceptance rate • Completed onboarding with no prior experience

Conducted data entry, document digitization, and annotated images and documents using CVAT and Label Studio for government and NGO projects. Achieved a 97% task acceptance rate and completed more than 40,000 micro-tasks during a six-month period. Developed early proficiencies in industry-standard annotation tools to meet client and project requirements.• Tagged and digitized images/documents • Used CVAT and Label Studio platforms • Maintained high task acceptance rate • Completed onboarding with no prior experience

2020 - 2020

Education

Work History