For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
N
Ntombizonke Z.

Ntombizonke Z.

LLM Data Labeling and Evaluation Specialist

South Africa flagDurban, South Africa

Key Skills

Software

LabelboxLabelbox
Other
Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code ProgrammingComputer Code Programming
ImageImage
VideoVideo

Top Task Types

Computer Programming/CodingComputer Programming/Coding
Point/Key PointPoint/Key Point
Prompt + Response Writing (SFT)Prompt + Response Writing (SFT)
Text GenerationText Generation
TranscriptionTranscription

Freelancer Overview

I have hands-on experience in AI training data, focusing on high-accuracy text and code annotation for large language models. My work includes LLM response evaluation, prompt review, ranking outputs, and identifying logical, factual, and formatting errors. I consistently follow detailed guidelines and maintain strict quality standards to ensure reliable training data. I’m detail-oriented, fast, and consistent. I adapt quickly to new task instructions, handle edge cases carefully, and deliver clean, defensible annotations. My background in analytical problem-solving helps me evaluate model outputs objectively and flag issues others miss. I prioritize accuracy over speed, but still meet deadlines without excuses.

Labeling Experience

RWS Multi-Image LLM Evaluation & Annotation

OtherImageImageBounding BoxBounding BoxPoint/Key PointPoint/Key Point

Worked on RWS multi-image LLM evaluation tasks requiring reasoning across multiple related images paired with text prompts. Assessed whether model responses correctly interpreted visual content across image sets, including objects, attributes, relationships, comparisons, and contextual details. Identified hallucinations, missed visual elements, incorrect associations between images, and flawed multi-step reasoning. Ranked multiple responses, flagged visual and logical errors, and applied strict evaluation rubrics to ensure alignment with instructions. Maintained high accuracy, consistency, and attention to detail on complex multimodal tasks under tight guidelines.

2025
Labelbox

LLM Text & Code Evaluation and Data Labeling

LabelboxLabelboxComputer Code ProgrammingComputer Code ProgrammingComputer Programming/CodingComputer Programming/Coding

Worked on code-focused LLM evaluation projects involving systematic review and annotation of model-generated responses. Tasks included assessing correctness, logical consistency, formatting, and adherence to instructions across programming-related prompts. Identified bugs, incomplete logic, hallucinations, and edge-case failures in generated code and explanations. Provided structured feedback, ranked multiple outputs, and applied strict evaluation rubrics to support model improvement. Maintained high attention to detail, followed evolving guidelines accurately, and delivered consistent, defensible annotations under time constraints. Focused on quality and precision rather than guessing or over-scoring.

2025

Education

U

University of KwaZulu-Natal

Bachelor of Science, Mathematics and Statistics

Bachelor of Science
2025