Adeyemi P.

Key Skills

Software

Don't disclose

Other

Deep Systems

Google Cloud Vertex AI

Micro1

AWS SageMaker

Top Subject Matter

Educational content evaluation for WAEC/JAMB/A-Level equivalent exams

Cybersecurity education and learner reasoning assessment

FinTech analytics (stock/forex datasets) and reporting

Top Data Types

Text

Document

Image

Top Task Types

Question Answering

Prompt + Response Writing (SFT)

Text Generation

Text Summarization

Fine-tuning

Transcription

Data Collection

Object Detection

Freelancer Overview

Digital Tutor & AI-Assisted Content Evaluator, Teesas Education. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose and Other. Education includes Bachelor of Engineering, Federal University of Agriculture, Abeokuta (FUNAAB) (2023). AI-training focus includes data types such as Text and Document and labeling workflows including Evaluation, Rating, and Question Answering.

Labeling Experience

Cybersecurity Tutor, Analytica Haven

Don't discloseText

Assessed student reasoning quality and delivered structured, evidence-based feedback identifying conceptual gaps. Used analytical reasoning and precise technical communication to support AI-like evaluation and improvement of responses. Translated dense technical concepts into clear, structured explanations for diverse learner cohorts.• Evaluated reasoning quality to identify conceptual gaps.• Produced structured feedback suitable for iterative improvement workflows.• Explained technical concepts clearly for varied learner groups.• Applied analytical reading/assessment skills aligned to LLM output evaluation.

2025 - Present

Prompt Engineering Case Study, Educational Content Pipeline 2025

Don't discloseTextPrompt + Response Writing (SFT)

Applied structured prompt engineering to improve AI-generated educational content across WAEC, JAMB, and A-Level equivalent materials. Evaluated before/after prompt comparisons to assess impacts on reasoning coherence, factual accuracy, and audience appropriateness. Adjusted prompt structure to guide model behavior toward more reliable instructional outputs.• Designed prompt structures for exam-aligned educational content.• Conducted before/after comparisons for prompt effectiveness.• Measured changes in reasoning coherence and factual accuracy.• Ensured audience fit via tone and difficulty adjustments.

2025 - 2025

LLM Failure Pattern Documentation, Teesas Education 2025

Don't discloseText

Documented recurring failure modes in LLM-generated educational content, including hallucinated facts, difficulty miscalibration, vague explanations, and tone misalignment. Implemented iterative prompt interventions to correct identified patterns and tracked output improvement across revision cycles. Repeated evaluation focused on reducing issues before content deployment.• Classified failure modes found in reviewed LLM outputs.• Applied prompt changes to address hallucinations and ambiguity.• Tracked improvement across multiple revision cycles.• Used structured documentation to monitor model behavior over time.

2025 - 2025

AI-Assisted Reporting & Human-in-the-Loop Evaluation (Contract), EISS 2025

Don't discloseDocumentQuestion Answering

Performed human-in-the-loop evaluation of AI-assisted report drafts to ensure factual accuracy and appropriate reasoning. Reviewed LLM outputs across a pipeline of operational reports to identify reasoning gaps and correct misaligned content. Ensured final deliverables met real-world business standards through documented evaluation decisions.• Used LLM-generated drafts as input to a multi-stage review pipeline.• Verified factual accuracy and corrected hallucinations or inaccuracies.• Identified reasoning gaps and corrected misaligned outputs.• Applied documented judgment to determine when to trust or intervene.

2025 - 2025

Digital Tutor & AI-Assisted Content Evaluator, Teesas Education

Don't discloseText

Evaluated LLM-generated educational content for factual accuracy, reasoning quality, coherence, and difficulty alignment before deployment. Performed human-in-the-loop QA by reviewing outputs daily and correcting misaligned outputs to improve instructional clarity. Documented recurring failure patterns and created revision workflows to reduce hallucinations, vague explanations, and difficulty miscalibration.• Reviewed 100+ LLM-generated pieces monthly for education pipelines.• Applied iterative prompt engineering to refine response quality and instructional clarity.• Identified failure modes (hallucinated facts, reasoning gaps, tone/difficulty misalignment).• Implemented systematic revision workflows based on documented patterns.

2024 - 2025

Show more

Education

Work History