Chibuzor Okafor

Key Skills

Software

CVAT

Labelbox

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Top Label Types

Bounding Box

Entity Ner Classification

Prompt Response Writing SFT

Freelancer Overview

I am an experienced full stack developer with a strong background in building and optimizing large-scale web and mobile applications, which has given me a deep understanding of data structures, data quality, and the importance of well-annotated datasets for AI-driven solutions. My work leading teams and collaborating with stakeholders has honed my attention to detail—essential for high-quality data labeling and annotation. I have hands-on experience with SQL, PostgreSQL, MongoDB, and DynamoDB, enabling me to manage and preprocess large datasets efficiently. Through projects in e-commerce and fintech, I have worked with real-time data processing, secure data handling, and automation, all of which are critical for preparing accurate training data for machine learning and AI applications. My familiarity with Python, RESTful APIs, and cloud platforms like AWS further supports my ability to contribute to AI training data initiatives, ensuring data integrity and scalability throughout the annotation process.

Labeling Experience

Prompt + Response Writing for Chatbot Fine-Tuning

Scale AITextPrompt Response Writing SFT

Designed structured prompt-response pairs for LLM supervised fine-tuning tasks. Covered classifications, reasoning problems, conversational tasks, knowledge Q&A, rewrites, safety filtering, and summarization. Generated 10,000+ high-quality pairs used for chatbot alignment, dialog flow improvements, and safety tuning. All responses adhered to tone, safety, and formatting guidelines.

Designed structured prompt-response pairs for LLM supervised fine-tuning tasks. Covered classifications, reasoning problems, conversational tasks, knowledge Q&A, rewrites, safety filtering, and summarization. Generated 10,000+ high-quality pairs used for chatbot alignment, dialog flow improvements, and safety tuning. All responses adhered to tone, safety, and formatting guidelines.

2024

Autonomous Vehicle Object Detection Dataset

CVATImageBounding Box

Labeled over 80,000 high-resolution traffic scene images for a self-driving car perception model. Annotated vehicles, pedestrians, signage, cyclists, and road obstacles using bounding boxes. Ensured accuracy through cross-review validation, IoU accuracy checks, and consistency audits. Dataset contributed to training an object detection model for lane navigation, obstacle detection, and pedestrian safety.

Labeled over 80,000 high-resolution traffic scene images for a self-driving car perception model. Annotated vehicles, pedestrians, signage, cyclists, and road obstacles using bounding boxes. Ensured accuracy through cross-review validation, IoU accuracy checks, and consistency audits. Dataset contributed to training an object detection model for lane navigation, obstacle detection, and pedestrian safety.

2023 - 2023

Medical Named Entity Recognition (NER)

LabelboxTextEntity Ner Classification

Annotated clinical notes, prescriptions, and discharge summaries to identify medical entities such as symptoms, diagnoses, medications, and procedures. Followed strict annotation guidelines for overlapping entities & multi-label cases. Dataset was used to train an NLP model assisting doctors with automated chart structuring and diagnosis suggestion.

Annotated clinical notes, prescriptions, and discharge summaries to identify medical entities such as symptoms, diagnoses, medications, and procedures. Followed strict annotation guidelines for overlapping entities & multi-label cases. Dataset was used to train an NLP model assisting doctors with automated chart structuring and diagnosis suggestion.

2022 - 2023

Education

Work History