Basil Osikuku

Key Skills

Software

Other

Appen

iMerit

Remotasks

Top Subject Matter

Natural Language Processing (NLP)

Large Language Models (LLMs)

AI/Data training

Top Data Types

Image

Video

Text

Top Task Types

RLHF

Classification

Bounding Box

Transcription

Evaluation Rating

Data Collection

Freelancer Overview

AI Data Annotation Intern. Core strengths include Other and Appen. Education includes Bachelor of Science, University of Toronto (2024). AI-training focus includes data types such as Text and Image and labeling workflows including RLHF, Classification, and Bounding Box.

Labeling Experience

AI Data Annotation Intern

OtherTextRLHF

As an AI Data Annotation Intern at DataForce by TransPerfect, I labeled and reviewed over 50,000 text data points in NLP datasets. My work involved flagging model hallucinations, ambiguous prompts, and ensuring adherence to RLHF-based annotation guidelines. High-quality collaboration with QA leads was essential to maintaining inter-annotator agreement rates of over 92% across multiple projects. • Labeled large volumes of text data for natural language understanding. • Documented edge cases and model errors to improve annotation accuracy. • Applied structured RLHF methodologies for consistent data quality. • Maintained detailed records to support project traceability and reproducibility.

As an AI Data Annotation Intern at DataForce by TransPerfect, I labeled and reviewed over 50,000 text data points in NLP datasets. My work involved flagging model hallucinations, ambiguous prompts, and ensuring adherence to RLHF-based annotation guidelines. High-quality collaboration with QA leads was essential to maintaining inter-annotator agreement rates of over 92% across multiple projects. • Labeled large volumes of text data for natural language understanding. • Documented edge cases and model errors to improve annotation accuracy. • Applied structured RLHF methodologies for consistent data quality. • Maintained detailed records to support project traceability and reproducibility.

2024 - 2024

Research Assistant – NLP Lab

OtherTextClassification

As a Research Assistant with the University of Toronto NLP Lab, I built a benchmark dataset of over 10,000 human-labeled sentence pairs for semantic similarity research. I was responsible for leading data annotation generations and implementing pre-processing pipelines for structured and unstructured data. This work contributed critical data for model evaluation and NLP algorithm development. • Assisted in annotation and verification of semantic labels on sentence pairs. • Implemented data preprocessing using Python. • Ensured data integrity through systematic QA and literature review. • Contributed to research supporting advances in semantic similarity computing.

As a Research Assistant with the University of Toronto NLP Lab, I built a benchmark dataset of over 10,000 human-labeled sentence pairs for semantic similarity research. I was responsible for leading data annotation generations and implementing pre-processing pipelines for structured and unstructured data. This work contributed critical data for model evaluation and NLP algorithm development. • Assisted in annotation and verification of semantic labels on sentence pairs. • Implemented data preprocessing using Python. • Ensured data integrity through systematic QA and literature review. • Contributed to research supporting advances in semantic similarity computing.

2023 - 2024

Freelance Data Labeller

AppenImageBounding Box

As a Freelance Data Labeller for Appen and Scale AI, I completed annotation tasks spanning image classification, text categorization, bounding box labeling, and named entity recognition. My projects included dialogue dataset annotation for conversational AI and RLHF preference ranking exercises. I ensured the usefulness and safety of AI models by rating outputs on helpfulness, harmlessness, and honesty. • Annotated images using bounding box tools for object detection. • Labeled text for categorization and entity recognition tasks. • Performed RLHF rankings and model output evaluations. • Actively contributed to AI model improvements in multiple domains.

As a Freelance Data Labeller for Appen and Scale AI, I completed annotation tasks spanning image classification, text categorization, bounding box labeling, and named entity recognition. My projects included dialogue dataset annotation for conversational AI and RLHF preference ranking exercises. I ensured the usefulness and safety of AI models by rating outputs on helpfulness, harmlessness, and honesty. • Annotated images using bounding box tools for object detection. • Labeled text for categorization and entity recognition tasks. • Performed RLHF rankings and model output evaluations. • Actively contributed to AI model improvements in multiple domains.

2022 - 2023

Education

Work History