Chimaobi J.

Key Skills

Software

Scale AI

Label Studio

Appen

Top Subject Matter

Healthcare AI - Medical Records & Patients Data

Live E-commerce Marketplace - User Experiences & Workflows

Fashion Design Illustration - Apparel Design & Editorials

Top Data Types

Text

Image

Video

Top Task Types

Prompt + Response Writing (SFT)

Function Calling

Evaluation/Rating

Freelancer Overview

With 6+ years in brand, UX, and product design across healthtech, fintech, and agro-allied sectors, I bring sharp editorial judgment and structured analytical thinking to AI training and data annotation work. My most direct experience comes from a contract engagement at Burna AI — a clinical AI platform — where I evaluated model-generated outputs for factual accuracy, hallucination risk, and tonal appropriateness, and designed feedback interfaces that helped align AI behavior with clinical user intent. Beyond clinical AI, I have applied annotation-adjacent skills across multiple projects: writing and iterating on prompts for LLMs, ranking and comparing model outputs, red-teaming edge cases in AI-generated UX copy, and building structured content pipelines that produce consistent, reviewable outputs. I hold a certification in AI Data Annotation & Labeling and bring domain fluency in healthcare, e-commerce, and brand/marketing contexts to every labeling task.

Labeling Experience

Product Designer — Live Commerce Platform (2025 – Present)

TextFunction Calling

Created structured prompt templates intended for AI-generated product listing copy and checkout content. Validated UX and edge cases using evaluation logic analogous to data labeling quality review. Supported consistent, context-appropriate AI output behavior through systematic testing of prompt-driven flows. • Developed prompt templates for product listing and seller onboarding content • Evaluated checkout UX against adversarial and edge-case inputs • Applied quality-review evaluation logic similar to labeling workflows • Ensured consistent AI output alignment with user context

2025 - Present

Video Generation, Motion Design & Content Annotation Enablement — Burna AI (Feb 2026 – Apr 2026)

OtherVideoEvaluation/Rating

• Built a prompt-driven motion graphics pipeline to generate reproducible content at scale. Used code automation to translate prompt inputs into structured outputs suitable for downstream use. Supported content annotation and taxonomy-style organization for repeatable generation workflows. • Reviewed AI-generated CTCAE v6.0 summaries for factual accuracy, hallucination risk, and appropriate clinical tone • Ranked and rated model response quality across multiple output variants to support preference-based alignment • Red-teamed edge cases in oncology workflows — identifying failure modes in structured AI outputs and writing corrective prompts • Designed annotation feedback surfaces enabling clinical staff to flag and score model outputs directly within the product UI • Built prompt-driven content pipelines via Remotion and Claude Code to produce structured, consistently formatted outputs for downstream annotation and review • Implemented automated, reproducible generation using Remotion and Claude Code • Created prompt-driven pipelines for consistent output behavior • Enabled scalable content generation aligned with labeling needs • Structured outputs for downstream annotation and review

2026 - 2026

AI ANNOTATION & Output Evaluation — Burna AI Clinical AI Platform (Feb 2026 – Apr 2026)

TextPrompt + Response Writing (SFT)

Embedded as Product & UX Designer on a clinical AI platform automating adverse event (AE) grading for cancer centers and clinical trials. Reviewed AI-generated CTCAE v6.0 summaries for factual accuracy, hallucination risk, and appropriate clinical tone Ranked and rated model response quality across multiple output variants to support preference-based alignment Red-teamed edge cases in oncology workflows — identifying failure modes in structured AI outputs and writing corrective prompts Designed annotation feedback surfaces enabling clinical staff to flag and score model outputs directly within the product UI

2026 - 2026

Education

Work History