Ezekiel A.

Key Skills

Software

Argilla

CVAT

CrowdSource

Top Subject Matter

LLM evaluation

prompt quality

reasoning assessment

Top Data Types

Text

Computer Code Programming

Image

Top Task Types

Prompt + Response Writing (SFT)

RLHF

Text Generation

Object Detection

Fine-tuning

Text Summarization

Question Answering

Computer Programming/Coding

Freelancer Overview

AI Trainer / LLM Evaluator (training project work & evaluation role). Brings 10+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose. Education includes Bachelor of Engineering, Federal University of Technology, Akure (FUTA) (2024) and Full-Stack Developer Bootcamp Certificate, 01-Edu / Learn2Earn (Yaba HQ) (2025). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

Data Annotation (RLHF/Preference data)

Don't discloseTextRLHF

Performed data annotation work for reinforcement learning from human feedback (RLHF) by labeling and ranking AI outputs. The annotation focused on preferences intended for RLHF and preference data pipelines. Quality decisions were driven by evaluation of response quality and alignment signals. •Labeled and ranked AI outputs by preference •Prepared preference data for RLHF pipelines •Assessed alignment, coherence, and reasoning quality •Used evaluation judgments to support preference learning

2025

Reasoning & Problem Creator (AI training focus)

Don't disclose

Created reasoning and problem sets for LLM practice, focusing on structured coding and logic challenges with step-by-step solution verification. The training emphasized problem decomposition and clean, readable reasoning workflows. These tasks were intended to improve LLM reasoning behavior and solution reliability. •Authored structured logic/coding challenges •Designed problems with step-by-step verification •Applied decomposition to build reasoning tasks •Used algorithmic thinking to validate solutions

2025

Prompt Engineer (AI training focus)

Don't discloseTextPrompt + Response Writing (SFT)

Performed prompt engineering by designing, testing, and refining prompts to elicit precise, high-quality model outputs. The training focused on crafting prompts that improved LLM quality and alignment with task requirements. Prompt iterations were validated through downstream evaluation of response quality. •Created prompts for better instruction-following •Iterated prompt wording to improve output precision •Tested prompts and validated response quality •Aligned outputs with requested reasoning/task constraints

2025

AI Trainer / LLM Evaluator (training project work & evaluation role)

Don't disclose

Trained as an LLM evaluator by assessing AI-generated outputs for accuracy, coherence, reasoning quality, and instruction alignment. This included judging whether model explanations were correct, clear, and helpful, and identifying flawed reasoning. The work emphasized structured reasoning checks and quality assessment of responses intended to improve LLM performance. •Evaluated AI response quality and adherence to prompts •Assessed reasoning correctness, clarity, and usefulness •Verified structured explanations against expected logic •Performed quality/alignment checks to support improved LLM outputs

2025

Education

Work History