Emmanuel O.

Key Skills

Software

Micro1

Appen

Mercor

OneForma

Labelbox

Mindrift

Top Subject Matter

Multilingual transcription and phonetic/linguistic labeling for speech-to-text datasets

RLHF-style response evaluation and gold-standard reference creation for LLMs

Legal Services & Contract Review

Top Data Types

Audio

Text

Document

Top Task Types

Transcription

RLHF

Freelancer Overview

I am a meticulous Data Annotation Specialist with extensive experience in RLHF (Reinforcement Learning from Human Feedback) and high-fidelity transcription. My work centers on refining LLM outputs by evaluating response accuracy, identifying model hallucinations, and providing logically rigorous justifications for model ranking. I have a proven track record in projects like MSPO 1654, where I utilized the uLabel platform to perform complex linguistic labeling with a consistent accuracy rating of 95%+, strictly adhering to multi-layered style guides. What sets me apart is my ability to bridge the gap between technical data requirements and linguistic nuance. With a Second Class Upper degree in Mass Communication, I excel at identifying subtle semantic errors and cultural edge cases that automated systems often miss. I am deeply committed to data integrity and possess the analytical stamina required to manage high-volume, complex annotation tasks without compromising on quality. AI-training focus includes data types such as Audio and Text and labeling workflows including Transcription and RLHF.

Labeling Experience

Linguistic Data Refinement (Independent AI Training)

TextRLHF

Conducted independent AI training by comparing LLM outputs and labeling responses for factual truthfulness, helpfulness, and harmlessness. Investigated and flagged edge cases where models missed cultural nuances or local context, with emphasis on West African linguistic structures. Authored gold-standard responses used to calibrate automated evaluation scripts. • Labeled model responses along RLHF criteria: truthfulness, helpfulness, and harmlessness • Identified failure modes and documented edge cases related to culture and local context • Produced gold-standard reference answers for evaluation calibration • Supported evaluation workflows with structured, logically consistent labeling outputs

Not specified

Multilingual Transcription & Labeling Specialist | Project MSPO 1654

AudioTranscription

Provided high-accuracy multilingual transcription and linguistic labeling to improve speech-to-text performance. Used uLabel to categorize phonetic variations while ensuring semantic alignment between English and regional dialects. Achieved consistent accuracy ratings of 95%+ while meeting daily throughput and strict 50+ page formatting requirements. • Transcribed multilingual audio verbatim with clean reads aligned to style guides • Labeled phonetic variation categories to support downstream speech-to-text modeling • Ensured semantic alignment across English and regional dialect utterances • Maintained high inter-activity accuracy under rigorous daily throughput targets

Not specified

Education

Work History