Idomo Mercy

Key Skills

Software

CVAT

Labelbox

AWS SageMaker

Other

Top Subject Matter

Computer Vision

Multimodal AI

Autonomous Driving

Top Data Types

Image

Video

Text

Audio

Document

Top Task Types

Bounding Box

Segmentation

Polygon

Point/Key Point

Object Detection

Tracking

Transcription

Entity (NER) Classification

Classification

Freelancer Overview

Data Annotation Intern — DataTag Solutions. Core strengths include CVAT, Labelbox, and AWS SageMaker. Education includes Bachelor of Science, University of London. AI-training focus includes data types such as Image, Video, and Text and labeling workflows including Bounding Box, Segmentation, and Polygon.

Labeling Experience

Data Annotation Intern — DataTag Solutions

CVATImageBounding BoxSegmentation

As a Data Annotation Intern at DataTag Solutions, I led annotation and QA for over 25 computer vision and multimodal projects. I designed and enforced annotation guidelines, built labeling pipelines, and partnered with ML engineers to perform dataset relabeling that improved model performance. I also trained and mentored junior annotators and managed data versioning for model reproducibility. • Produced and validated 240,000+ annotations for images, videos, and audio data. • Designed annotation schema and improved guidelines, boosting inter-annotator agreement and reducing disputes. • Automated preprocessing using CVAT and Python, leading to a 50% reduction in manual work and a 35% increase in throughput. • Standardized onboarding and QA, cutting ramp-up time for new annotators and increasing project efficiency.

As a Data Annotation Intern at DataTag Solutions, I led annotation and QA for over 25 computer vision and multimodal projects. I designed and enforced annotation guidelines, built labeling pipelines, and partnered with ML engineers to perform dataset relabeling that improved model performance. I also trained and mentored junior annotators and managed data versioning for model reproducibility. • Produced and validated 240,000+ annotations for images, videos, and audio data. • Designed annotation schema and improved guidelines, boosting inter-annotator agreement and reducing disputes. • Automated preprocessing using CVAT and Python, leading to a 50% reduction in manual work and a 35% increase in throughput. • Standardized onboarding and QA, cutting ramp-up time for new annotators and increasing project efficiency.

2021 - 2025

Data Annotation Intern — LabelNest UK

LabelboxImageBounding BoxSegmentation

At LabelNest UK, I executed high-volume labeling projects across image, text, and audio domains. I created annotation specification documents, implemented batch QA processes, and converted legacy annotation formats for standardized pipelines. I conducted inter-annotator agreement studies to help prioritize guideline updates and improve dataset quality. • Labeled 150,000+ images, 30,000+ sentence-level text samples, and QA for 200+ hours of audio transcription. • Developed and refined annotation guidelines for NER and other tasks, improving benchmark F1 scores. • Built QA dashboards and error-tracking using Excel and SQL, reducing rework volume by 30%. • Standardized dataset formats using Python scripts for downstream machine learning use.

At LabelNest UK, I executed high-volume labeling projects across image, text, and audio domains. I created annotation specification documents, implemented batch QA processes, and converted legacy annotation formats for standardized pipelines. I conducted inter-annotator agreement studies to help prioritize guideline updates and improve dataset quality. • Labeled 150,000+ images, 30,000+ sentence-level text samples, and QA for 200+ hours of audio transcription. • Developed and refined annotation guidelines for NER and other tasks, improving benchmark F1 scores. • Built QA dashboards and error-tracking using Excel and SQL, reducing rework volume by 30%. • Standardized dataset formats using Python scripts for downstream machine learning use.

2018 - 2021

Annotation Intern (Contract) — AnnotaCore Ltd

OtherAudioTranscriptionBounding Box

During my contract at AnnotaCore Ltd, I worked on short-term projects focused on labeling for academic and startup needs. Tasks included audio transcription and speaker diarization, medical image annotation, and preparing text corpora for annotation. I developed QA checklists and automated data cleaning, ensuring rapid delivery for aggressive project timelines. • Labeled 100+ hours of audio for transcription and speaker identification tasks. • Annotated 5,000+ medical images with bounding boxes and segmentation labels. • Prepared and cleaned large-scale text datasets using Python (pandas, regex) for downstream annotation. • Built lightweight reporting documentation and QA processes for project traceability.

During my contract at AnnotaCore Ltd, I worked on short-term projects focused on labeling for academic and startup needs. Tasks included audio transcription and speaker diarization, medical image annotation, and preparing text corpora for annotation. I developed QA checklists and automated data cleaning, ensuring rapid delivery for aggressive project timelines. • Labeled 100+ hours of audio for transcription and speaker identification tasks. • Annotated 5,000+ medical images with bounding boxes and segmentation labels. • Prepared and cleaned large-scale text datasets using Python (pandas, regex) for downstream annotation. • Built lightweight reporting documentation and QA processes for project traceability.

2017 - 2018

Education

Work History