For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
O

Okikioluwa Karunwi

AI Trainer

Nigeria flagN/A, Nigeria
$35.00/hrExpertAppenScale AIDon T Disclose

Key Skills

Software

AppenAppen
Scale AIScale AI
Don't disclose

Top Subject Matter

Healthcare Data & Medical Documentation
Science Domain Expertise
General Knowledge

Top Data Types

TextText
ImageImage
AudioAudio

Top Task Types

RLHFRLHF
ClassificationClassification
Entity (NER) ClassificationEntity (NER) Classification
SegmentationSegmentation
Text SummarizationText Summarization
Text GenerationText Generation
Question AnsweringQuestion Answering
Prompt + Response Writing (SFT)Prompt + Response Writing (SFT)
Object DetectionObject Detection
TranscriptionTranscription
Data CollectionData Collection

Freelancer Overview

AI Trainer — RLHF & Model Evaluation. Brings 3+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Internal and Proprietary Tooling. Education includes Master of Science, Rome Business School (2024) and Bachelor of Science, Obafemi Awolowo University (2019). AI-training focus includes data types such as Text, Medical, and DICOM and labeling workflows including RLHF, Classification, and Entity (NER) Classification.

ExpertEnglish

Labeling Experience

AI Trainer — RLHF & Model Evaluation

TextRLHF
As an AI Trainer focusing on RLHF and model evaluation, I assessed AI-generated outputs for accuracy, coherence, instruction-following, and safety across diverse domains. I engineered and tested prompts relevant to healthcare, science, and general knowledge, ranking responses to identify and document edge cases. I delivered a comprehensive conversational AI dataset that achieved high annotation quality and informed improvements in annotation schema. • Led the RLHF feedback loop for fine-tuning large language models. • Identified hallucinations and categorized reasoning/safety failures to enhance annotation protocols. • Produced a 120,000-token intent classification, slot filling, and dialogue act labeling dataset. • Achieved inter-annotator agreement of Kappa 0.93, surpassing the set benchmark.

As an AI Trainer focusing on RLHF and model evaluation, I assessed AI-generated outputs for accuracy, coherence, instruction-following, and safety across diverse domains. I engineered and tested prompts relevant to healthcare, science, and general knowledge, ranking responses to identify and document edge cases. I delivered a comprehensive conversational AI dataset that achieved high annotation quality and informed improvements in annotation schema. • Led the RLHF feedback loop for fine-tuning large language models. • Identified hallucinations and categorized reasoning/safety failures to enhance annotation protocols. • Produced a 120,000-token intent classification, slot filling, and dialogue act labeling dataset. • Achieved inter-annotator agreement of Kappa 0.93, surpassing the set benchmark.

2023 - Present

Annotation Schema Designer — Conversational AI Client

TextEntity Ner Classification
For a Conversational AI client, I designed and implemented a multi-label ontology annotation schema across a complex NLP pipeline. My schema covered numerous intent categories and slot types, leading to substantial gains in inter-annotator agreement. This schema was adopted as a client-wide standard for future model iterations. • Solely responsible for annotation schema design in a 12-project NLP pipeline. • Enabled 20-person team to achieve Kappa 0.93 inter-annotator agreement. • Standardized multi-label ontology for intent and slot annotation. • Schema adopted by client for all subsequent model deployments.

For a Conversational AI client, I designed and implemented a multi-label ontology annotation schema across a complex NLP pipeline. My schema covered numerous intent categories and slot types, leading to substantial gains in inter-annotator agreement. This schema was adopted as a client-wide standard for future model iterations. • Solely responsible for annotation schema design in a 12-project NLP pipeline. • Enabled 20-person team to achieve Kappa 0.93 inter-annotator agreement. • Standardized multi-label ontology for intent and slot annotation. • Schema adopted by client for all subsequent model deployments.

2023 - 2023

Medical Image Annotation Specialist — Healthcare AI Client

Segmentation
For a Healthcare AI client, I delivered pixel-level semantic segmentation on thousands of chest X-ray images. I collaborated with radiologists to validate labeling, and developed a labeling schema that ensured testing accuracy. My work achieved a high Dice coefficient and was distinct from broader computer vision pipeline tasks. • Performed semantic segmentation on 8,500 chest X-rays. • Developed the annotation schema in collaboration with subject matter experts. • Managed QA processes for all annotation rounds. • Achieved a Dice coefficient of 0.91 on the held-out dataset.

For a Healthcare AI client, I delivered pixel-level semantic segmentation on thousands of chest X-ray images. I collaborated with radiologists to validate labeling, and developed a labeling schema that ensured testing accuracy. My work achieved a high Dice coefficient and was distinct from broader computer vision pipeline tasks. • Performed semantic segmentation on 8,500 chest X-rays. • Developed the annotation schema in collaboration with subject matter experts. • Managed QA processes for all annotation rounds. • Achieved a Dice coefficient of 0.91 on the held-out dataset.

2022 - 2022

Data Annotation Specialist

TextClassification
As a Data Annotation Specialist, I labeled over 200,000 data points spanning text, images, and audio, with a focus on precise and quality annotations. I created comprehensive annotation guidelines and ontologies for multiple projects, reducing disagreement among teams. I produced named entity recognition, relation extraction, sentiment classification, segmentation masks, bounding box, and keypoint labels that improved downstream model performance. • Maintained a QA-audited 99.2% annotation accuracy across all modalities. • Led annotation efforts contributing to F1-score improvements on multiple AI models. • Produced image annotations for healthcare and retail computer vision projects. • Designed processes that reduced inter-annotator disagreement by 38%.

As a Data Annotation Specialist, I labeled over 200,000 data points spanning text, images, and audio, with a focus on precise and quality annotations. I created comprehensive annotation guidelines and ontologies for multiple projects, reducing disagreement among teams. I produced named entity recognition, relation extraction, sentiment classification, segmentation masks, bounding box, and keypoint labels that improved downstream model performance. • Maintained a QA-audited 99.2% annotation accuracy across all modalities. • Led annotation efforts contributing to F1-score improvements on multiple AI models. • Produced image annotations for healthcare and retail computer vision projects. • Designed processes that reduced inter-annotator disagreement by 38%.

2022 - 2022

Data Integrity & QA Analyst

Classification
While serving as a Data Integrity & QA Analyst, I audited and corrected clinical records with a focus on ICD-10 coding. My efforts improved the accuracy and completeness of healthcare datasets, directly supporting medical AI training protocols. I trained junior annotators to meet rigorous quality standards and reduced rework rates on labeled data. • Processed and corrected 30,000+ clinical records for AI applications. • Achieved a peer-reviewed annotation accuracy of 98.7% on medical codes. • Developed enrichment protocols, increasing dataset completeness by 29%. • Mentored a junior annotation team to meet healthcare QA requirements.

While serving as a Data Integrity & QA Analyst, I audited and corrected clinical records with a focus on ICD-10 coding. My efforts improved the accuracy and completeness of healthcare datasets, directly supporting medical AI training protocols. I trained junior annotators to meet rigorous quality standards and reduced rework rates on labeled data. • Processed and corrected 30,000+ clinical records for AI applications. • Achieved a peer-reviewed annotation accuracy of 98.7% on medical codes. • Developed enrichment protocols, increasing dataset completeness by 29%. • Mentored a junior annotation team to meet healthcare QA requirements.

2021 - 2021

Education

O

Obafemi Awolowo University

Bachelor of Science, Microbiology / Public Health

Bachelor of Science
2014 - 2019
R

Rome Business School

Master of Science, eHealth Management

Master of Science
2024

Work History

A

Academic Research Contracts

Research Data Analyst & Annotator

N/A
2019 - 2021