Quadri Busari

Key Skills

Software

CVAT

Label Studio

V7 Labs

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Audio

Document

Top Label Types

Bounding Box

Entity Ner Classification

Transcription

Freelancer Overview

I am an experienced AI trainer and data specialist with a multidisciplinary background in industrial chemistry, business strategy, and astrophysics. My work centers on generating high-quality training data, authoring complex prompts, and performing rigorous data labeling and annotation for large language models. I have hands-on expertise in Reinforcement Learning from Human Feedback (RLHF), chain-of-thought reasoning, adversarial testing, and ground truth generation, with a strong focus on reducing model hallucinations and improving factual accuracy in technical domains. My technical skills include Python, SQL, and advanced Excel, and I have applied these tools to process and model large datasets in scientific and business contexts. I excel at synthesizing complex information across domains—such as chemistry, business, and physics—to create nuanced AI training scenarios, optimize model logic, and ensure data integrity for advanced reasoning tasks.

Labeling Experience

Freelance Data Annotation Specialist – Audio Transcription & Labeling

Label StudioAudioTranscription

Transcribed and labeled over 500 hours of Nigerian English, Yoruba, and Pidgin audio for Automatic Speech Recognition (ASR) development. Provided timestamp-level accuracy and multi-speaker identification for diverse datasets across languages and dialects. Supported multilingual AI model training by delivering precise annotations and quality-audited transcripts. • Employed Label Studio, CVAT, and SuperAnnotate for audio data tasks • Ensured temporal consistency and accuracy in speaker diarization and emotion tagging • Created and refined guidelines for consistent audio labeling • Managed quality control and correction across large-scale transcription projects

Transcribed and labeled over 500 hours of Nigerian English, Yoruba, and Pidgin audio for Automatic Speech Recognition (ASR) development. Provided timestamp-level accuracy and multi-speaker identification for diverse datasets across languages and dialects. Supported multilingual AI model training by delivering precise annotations and quality-audited transcripts. • Employed Label Studio, CVAT, and SuperAnnotate for audio data tasks • Ensured temporal consistency and accuracy in speaker diarization and emotion tagging • Created and refined guidelines for consistent audio labeling • Managed quality control and correction across large-scale transcription projects

2022

Freelance Data Annotation Specialist – Text Entity Recognition

Label StudioTextEntity Ner Classification

Annotated more than 120,000 technical sentences for Named Entity Recognition (NER) projects within chemistry, business, and astronomical subjects. Maintained high inter-annotator agreement and precise taxonomy consistency across diverse categorization tasks. Supported entity extraction for advanced machine learning models by following strict guidelines. • Used Label Studio, Labelbox, and CVAT for text annotation projects • Tagged entities across multiple technical domains and languages as required • Developed detailed entity guidelines for consistency across annotators • Audited and corrected text labels for improved dataset accuracy

Annotated more than 120,000 technical sentences for Named Entity Recognition (NER) projects within chemistry, business, and astronomical subjects. Maintained high inter-annotator agreement and precise taxonomy consistency across diverse categorization tasks. Supported entity extraction for advanced machine learning models by following strict guidelines. • Used Label Studio, Labelbox, and CVAT for text annotation projects • Tagged entities across multiple technical domains and languages as required • Developed detailed entity guidelines for consistency across annotators • Audited and corrected text labels for improved dataset accuracy

2022

Freelance Data Annotation Specialist – Image Annotation

CVATImageBounding Box

Labeled over 80,000 scientific and industrial images using bounding boxes, polygons, and segmentation tools for computer vision model training. Ensured extremely high quality with a consistent accuracy rate above 99% and detailed taxonomy guidelines. Specialized in annotating equipment, apparatus, and scientific instrument images while collaborating remotely on global annotation projects. • Used CVAT, Labelbox, V7, SuperAnnotate, and Label Studio platforms • Handled edge cases and ambiguous images with expert-level quality notes • Performed large-scale quality audits, reviewing and correcting team annotations • Created custom taxonomies for multi-class and multi-tier image classification tasks

Labeled over 80,000 scientific and industrial images using bounding boxes, polygons, and segmentation tools for computer vision model training. Ensured extremely high quality with a consistent accuracy rate above 99% and detailed taxonomy guidelines. Specialized in annotating equipment, apparatus, and scientific instrument images while collaborating remotely on global annotation projects. • Used CVAT, Labelbox, V7, SuperAnnotate, and Label Studio platforms • Handled edge cases and ambiguous images with expert-level quality notes • Performed large-scale quality audits, reviewing and correcting team annotations • Created custom taxonomies for multi-class and multi-tier image classification tasks

2022

Data Analyst (Astronomy Focus) – Image Annotation

V7 LabsImageBounding Box

Labeled 10,000+ astronomical images by identifying and boxing galaxies, stars, nebulae, and artifacts for celestial object classification. Ensured sub-pixel annotation precision and clean dataset structuring for automated classification systems. Collaborated on large-scale astronomical data projects, processing both imagery and spectra. • Utilized CVAT, V7, and Labelbox tools for annotation • Marked diverse celestial object types and handled ambiguous cases expertly • Contributed to annotated training sets for astronomy machine learning models • Followed rigorous workflow for scientific accuracy in labeling

Labeled 10,000+ astronomical images by identifying and boxing galaxies, stars, nebulae, and artifacts for celestial object classification. Ensured sub-pixel annotation precision and clean dataset structuring for automated classification systems. Collaborated on large-scale astronomical data projects, processing both imagery and spectra. • Utilized CVAT, V7, and Labelbox tools for annotation • Marked diverse celestial object types and handled ambiguous cases expertly • Contributed to annotated training sets for astronomy machine learning models • Followed rigorous workflow for scientific accuracy in labeling

2022 - 2022

Business Strategy Consultant – Document Analysis & Tagging

Label StudioDocumentEntity Ner Classification

Processed and manually tagged over 800 business and legal documents for entity extraction, date tagging, and clause classification. Built and maintained document classification systems across 15 industrial sectors supporting structured data capture. Ensured high accuracy during extraction of data points and categorization for compliance and market analysis. • Used Label Studio and Excel for manual and semi-automated document annotation • Extracted and labeled key business, financial, and legal entities • Developed and documented multi-level document taxonomy structures • Maintained 99.8% accuracy for data extraction and labeling tasks

Processed and manually tagged over 800 business and legal documents for entity extraction, date tagging, and clause classification. Built and maintained document classification systems across 15 industrial sectors supporting structured data capture. Ensured high accuracy during extraction of data points and categorization for compliance and market analysis. • Used Label Studio and Excel for manual and semi-automated document annotation • Extracted and labeled key business, financial, and legal entities • Developed and documented multi-level document taxonomy structures • Maintained 99.8% accuracy for data extraction and labeling tasks

2014 - 2018

Education

Work History