For employers

Hire this AI Trainer

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

LoginCreate account
Invite to Job
Raphael Muselema

Raphael Muselema

AI Data Quality Analyst - Technical Logic

ZAMBIA flag
Kitwe, Zambia
$15.00/hrEntry LevelInternal Proprietary Tooling

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

TextText

Top Label Types

Prompt Response Writing SFT

Freelancer Overview

Electro-Mechanical Engineering graduate specializing in RLHF and technical logic verification for Large Language Models (LLMs). My core expertise lies in creating "Golden Set" ground-truth data used to benchmark model performance in STEM and procedural reasoning tasks. I differentiate myself through a rigorous engineering approach to data annotation, applying Boolean logic and Chain-of-Thought (CoT) verification to identify subtle hallucinations and edge cases that standard annotators miss. With a track record of annotating over 10,000 complex data points while maintaining a >98% quality score, I combine high-volume throughput with the precision required for Supervised Fine-Tuning (SFT). I am proficient in identifying logical fallacies in technical content and have experience defining Standard Operating Procedures (SOPs) for quality assurance, ensuring that training data meets strict schema requirements.

Entry LevelEnglish

Labeling Experience

AI Data Quality Analyst (Technical Logic Focus)

Internal Proprietary ToolingTextPrompt Response Writing SFT
Selected as a "Golden Set" author to generate high-fidelity ground truth data for Supervised Fine-Tuning (SFT). My role involves writing and evaluating complex technical prompts related to STEM reasoning, Boolean logic, and sequential instruction following. I conduct root cause analysis on model hallucinations and edge cases, maintaining a >98% acceptance rate against strict engineering-grade rubrics. I specifically focus on correcting logical fallacies in multi-turn technical conversations.

Selected as a "Golden Set" author to generate high-fidelity ground truth data for Supervised Fine-Tuning (SFT). My role involves writing and evaluating complex technical prompts related to STEM reasoning, Boolean logic, and sequential instruction following. I conduct root cause analysis on model hallucinations and edge cases, maintaining a >98% acceptance rate against strict engineering-grade rubrics. I specifically focus on correcting logical fallacies in multi-turn technical conversations.

2025

Education

T

The Copperbelt University

Bachelor of Engineering, Electromechanical Engineering

Bachelor of Engineering
2021 - 2025

Work History

P

Primegrade Solutions

Technical Content Coordinator

Kitwe
2024 - Present