Seyi P.

Key Skills

Software

Micro1

Don't disclose

Top Subject Matter

Structured data QA for AI training datasets (record validation, error tagging, guideline-based labeling)

Port/shipment operational analysis dataset labeling for modeling

RFM customer segmentation labeling for predictive modeling

Top Data Types

Text

Video

Top Task Types

Classification

Text Generation

Freelancer Overview

Data Validation and Quality Assurance Lead | Excelerate Analytics Team. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Google Colab, Excel, and GitHub. Education includes Bachelor of Science, Lagos State University (2028) and Data Analyst 101 Certification, TechCrush (2025). AI-training focus includes data types such as Text and labeling workflows including Classification.

Labeling Experience

Data Validation and Quality Assurance Lead | Excelerate Analytics Team

TextClassification

Led data validation and label quality assurance for large volumes of structured datasets, applying detailed annotation guidelines to flag structural errors, formatting inconsistencies, and logical discrepancies. Automated first-pass quality checks using Python scripts to tag problematic records with specific error categories before manual review. Maintained traceable annotation logs documenting each flagged record, error type, and resolution to ensure reproducibility across the dataset lifecycle. • Reviewed and annotated high-volume structured data batches for QA • Applied guideline interpretation to detect edge cases and inconsistencies • Documented labeling decisions and error resolutions for audit trail • Coordinated with a remote cross-functional team to align labeling standards and resolve ambiguities

2026 - 2026

AI Training Data Preparation — Apapa Port Analysis

TextClassification

Designed and structured a 5,000-record shipment dataset for AI training by defining classification categories, labeling schema, and annotation guidelines for delay types, lead time bands, and shipment statuses. Applied multi-class classification labels across all records using SQL-based logic and Excel validation tools to keep labels accurate and consistent with no ambiguous or conflicting entries. Documented the full labeling methodology, category definitions, and edge case decisions in a GitHub README to enable reproducibility and auditability. • Built shipment delay/lead-time status labels from scratch • Implemented multi-class labeling using SQL logic and Excel validation • Ensured dataset-wide consistency through validation checks • Produced documentation and edge case rationale for transparent reuse

2025 - 2026

Data Classification Simulation — Deloitte via Forage

TextClassification

Simulated rule-based data labeling for a multi-variable business dataset by applying structured classification logic to assign performance labels based on defined scoring criteria. Used advanced Excel functions to compute scores and translate them into the final labeled categories. Replicated workflows similar to AI data preparation steps used to create labeled training inputs. • Applied scoring rules to derive performance labels • Used multi-variable inputs for classification • Translated business criteria into label outputs • Practiced rule-based annotation workflow patterns for AI preparation

2025 - 2025

Agricultural Data Annotation and Categorization — Academic Collaborative Project

TextClassification

Annotated and categorized approximately 2,000 agricultural production records across four Nigerian states using consistent labels for loss type, cause category, and severity level. Resolved labeling inconsistencies across team members by standardizing category definitions and applying unified guidelines throughout the dataset. Ensured consistent classification outputs despite collaborative, multi-annotator workflows. • Labeled agricultural records with loss type, cause, and severity categories • Harmonized label definitions across annotators • Applied standardized guidelines to improve consistency • Identified and corrected cross-team labeling disagreements

2025 - 2025

Structured Data Labeling and Classification — E-commerce Transactions Project

TextClassification

Performed structured data labeling for an e-commerce transactions dataset by assigning RFM-based customer segment categories to each of 525,462 transaction records. Conducted multi-field data validation by cross-checking classification outputs against source data to identify and correct mislabeled entries prior to reporting. Maintained a documented annotation workflow including label definitions, scoring logic, and QA steps to make the process transparent and repeatable. • Labeled high-volume transactions using rule-based RFM classification • Validated outputs across multiple fields to prevent mislabeling • Implemented QA checks before final dataset reporting • Documented label definitions and scoring logic for consistency

2025 - 2025

Education

Work History