Ode J.

Key Skills

Software

Other

OneForma

Data Annotation Tech

Appen

Top Subject Matter

AI model output evaluation and structured labeling for factuality

Preference Domain Expertise

and relevance tasks (fintech/crypto-adjacent content)

Top Data Types

Text

Image

Audio

Top Task Types

Transcription

Classification

Object Detection

Question Answering

RLHF

Data Collection

Text Generation

Bounding Box

Evaluation/Rating

Segmentation

Freelancer Overview

Independent Annotation & Evaluation Contributor | Outlier, Appen, DataAnnotation, Prolific, OneForma. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Outlier and Other. Education includes Bachelor of Science, Miva Open University (2029). AI-training focus includes data types such as Text and Audio and labeling workflows including Evaluation, Rating, and Transcription.

Labeling Experience

Independent Annotation & Evaluation Contributor (Transcript Accuracy Reviews) | Outlier, Appen, DataAnnotation, Prolific, OneForma

OtherAudioTranscription

Evaluated transcript accuracy by checking consistency and identifying inaccuracies in AI-generated or provided transcriptions. Flagged inconsistencies and out-of-scope samples for further review, providing clear written explanations to support decisions. Maintained guideline adherence and documented findings to support reproducibility across review batches. • Reviewed transcripts for factual or structural inconsistencies • Marked low-confidence or out-of-scope items for escalation • Recorded clear explanations for each flagged sample • Followed platform-specific assessment and documentation requirements

2024 - Present

Independent Annotation & Evaluation Contributor | Outlier, Appen, DataAnnotation, Prolific, OneForma

Text

Served as an independent annotation and evaluation contributor for multiple crowdsourcing platforms, focusing on assessing and ranking AI model outputs against detailed guidelines. Performed factuality, reward-model preference, search relevance, and content quality evaluations, documenting decisions with evidence and written justifications. Escalated ambiguous, edge-case, and out-of-scope samples for review while maintaining reproducible documentation across high-volume batches. • Benchmarked biographies for unsupported or fabricated claims using source material • Ranked and rated competing model responses for reward model preference using rubrics • Rated query–result pairs for search relevance according to guideline definitions • Assessed social media ads and feed content quality/appropriateness with escalation reasoning

2024 - Present

Education

Work History