Paul Michuki

Key Skills

Software

Appen

CVAT

Data Annotation Tech

Dataloop

Google Cloud Vertex AI

Labelbox

Label Studio

Mercor

Playment

Roboflow

Scale AI

SuperAnnotate

Surge AI

Toloka

Telus

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Document

Image

Text

Video

Top Label Types

Action Recognition

Bounding Box

Classification

Computer Programming Coding

Object Detection

Polygon

Red Teaming

Text Generation

Transcription

Translation Localization

Freelancer Overview

I am an experienced AI Trainer and Data Annotator with over three years of hands-on experience delivering high-quality training data across text, audio, video, and image modalities. My background in Computer Science and Natural Language Processing enables me to excel in prompt engineering, AI model evaluation, and multi-modal data labeling, with a strong focus on accuracy, consistency, and adherence to complex guidelines. I am fluent in both English and Swahili, and have contributed to diverse NLP and multilingual AI projects, including adversarial testing, linguistic data validation, and cross-cultural model evaluation. My technical expertise includes advanced use of web-based annotation platforms, NLP tools, and quality assurance frameworks, and I am highly skilled in research, fact-checking, and systematic error identification. I thrive in remote, fast-paced environments, consistently meeting tight deadlines and adapting quickly to evolving project requirements.

Labeling Experience

AI Data Labeller & Prompt Engineer

Label StudioTextEvaluation Rating

As an AI Data Labeller & Prompt Engineer, I managed high-volume labeling projects across text, audio, image, and video. I evaluated AI-generated content for quality, safety, and relevance, adhering to strict annotation guidelines. My work included prompt engineering, bilingual validation, and adversarial testing to improve model robustness. • Expert in English and Swahili for bilingual/multilingual annotation • Maintained 98%+ annotation accuracy on large datasets • Applied systematic validation, error flagging, and iterative correction • Conducted fact-checking, source verification, and dataset quality improvement

As an AI Data Labeller & Prompt Engineer, I managed high-volume labeling projects across text, audio, image, and video. I evaluated AI-generated content for quality, safety, and relevance, adhering to strict annotation guidelines. My work included prompt engineering, bilingual validation, and adversarial testing to improve model robustness. • Expert in English and Swahili for bilingual/multilingual annotation • Maintained 98%+ annotation accuracy on large datasets • Applied systematic validation, error flagging, and iterative correction • Conducted fact-checking, source verification, and dataset quality improvement

2023

Image, Video & Audio Annotation Specialist

RoboflowImagePolygon

Annotated large-scale image datasets using bounding boxes, polygon segmentation, semantic segmentation, and keypoint labelling for object detection, pose estimation, and scene understanding tasks • Used Roboflow to build, manage, and version image annotation projects — including dataset augmentation, train/validation/test splits, and export to YOLO, COCO, and Pascal VOC formats • Applied YOLO v5 and v8 frameworks for annotation validation and model-assisted labelling, accelerating throughput on high-volume computer vision datasets • Worked in CVAT and LabelImg to annotate multi-class image datasets for retail, agriculture, medical imaging, and general object recognition use cases • Performed frame-by-frame video annotation for action recognition, object tracking, and activity detection — labelling object trajectories, temporal events, and scene transitions across long video sequences • Annotated video datasets for autonomous driving and surveillance use cases, including lane detection

Annotated large-scale image datasets using bounding boxes, polygon segmentation, semantic segmentation, and keypoint labelling for object detection, pose estimation, and scene understanding tasks • Used Roboflow to build, manage, and version image annotation projects — including dataset augmentation, train/validation/test splits, and export to YOLO, COCO, and Pascal VOC formats • Applied YOLO v5 and v8 frameworks for annotation validation and model-assisted labelling, accelerating throughput on high-volume computer vision datasets • Worked in CVAT and LabelImg to annotate multi-class image datasets for retail, agriculture, medical imaging, and general object recognition use cases • Performed frame-by-frame video annotation for action recognition, object tracking, and activity detection — labelling object trajectories, temporal events, and scene transitions across long video sequences • Annotated video datasets for autonomous driving and surveillance use cases, including lane detection

2021

AI Data Annotator & Quality Assurance Specialist

AppenAudioClassification

As an AI Data Annotator & Quality Assurance Specialist, I annotated and categorized audio, image, and text datasets for AI and NLP projects. I performed systematic data review, quality validation, and correction in multilingual contexts. Specialized Swahili language tasks and documentation of edge cases improved translation quality and AI usability. • Supported language technology projects through structured annotation cycles • Identified incorrect or culturally inappropriate model outputs • Collaborated with project teams for deadline adherence and data quality • Enhanced AI performance through detailed observation and remote work adaptation

As an AI Data Annotator & Quality Assurance Specialist, I annotated and categorized audio, image, and text datasets for AI and NLP projects. I performed systematic data review, quality validation, and correction in multilingual contexts. Specialized Swahili language tasks and documentation of edge cases improved translation quality and AI usability. • Supported language technology projects through structured annotation cycles • Identified incorrect or culturally inappropriate model outputs • Collaborated with project teams for deadline adherence and data quality • Enhanced AI performance through detailed observation and remote work adaptation

2021 - 2023

Education

Work History