Obuchinezia Anyanwu

Key Skills

Software

Scale AI

Other

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Task Types

Classification

Prompt + Response Writing (SFT)

Question Answering

RLHF

Freelancer Overview

I specialise in high-precision AI training data creation, with a focus on reasoning-heavy data labelling aligned with RLHF (Reinforcement Learning from Human Feedback) pipelines. As a top-performing contributor to one of Outlier’s key note projects, I have authored numerous expert-level prompts with clearly defined Ground Truth Final Answers (GTFA), optimised to challenge and train large language models in structured, multi-domain reasoning. My work regularly stumps frontier models and has been recognised as Master's and PhD-level by automated and peer grading systems. Beyond prompt creation, I’ve contributed to quality assurance by verifying logic paths and annotation consistency, and was selected for a rubric-building role where I’ll define evaluation criteria used to assess LLM outputs. My domain expertise spans medical affairs broadly, clinical trials, real-world evidence (RWE), regulatory science, and psychometrics, making me uniquely positioned to deliver training data that reflects both technical accuracy and real-world complexity.

Labeling Experience

Freelance Contributor – Expert Projects

Scale AITextQuestion AnsweringRLHF

Performed high-accuracy text data labeling by generating prompts with a single verifiable answer (GTFA) used in training and evaluating large language models (LLMs). Tasks involved complex reasoning annotation, including text categorization, logical consistency checks, and model response validation. Frequently assigned to domains such as clinical science, regulatory strategy, and real-world evidence (RWE), ensuring label accuracy in expert-level contexts. Also qualified for reviewer status, which involves validating peer-generated tasks, assessing reasoning quality, and verifying model stumps, further demonstrating proficiency in data quality assurance and alignment with RLHF objectives.

Performed high-accuracy text data labeling by generating prompts with a single verifiable answer (GTFA) used in training and evaluating large language models (LLMs). Tasks involved complex reasoning annotation, including text categorization, logical consistency checks, and model response validation. Frequently assigned to domains such as clinical science, regulatory strategy, and real-world evidence (RWE), ensuring label accuracy in expert-level contexts. Also qualified for reviewer status, which involves validating peer-generated tasks, assessing reasoning quality, and verifying model stumps, further demonstrating proficiency in data quality assurance and alignment with RLHF objectives.

2024

Education

Work History