Mmesoma Udensi

Designed and deployed an LLM-powered ETL pipeline using the Gemini API, transforming unstructured educational data into CTDL-compliant formats for 60+ partner institutions.

Built scalable ingestion tools with Python and Flask, reducing partner onboarding time by 80% and enabling seamless integration for non-technical users.

Implemented schema validation, mapping, and audit traceability, ensuring data accuracy and regulatory compliance across data processes.

Led Agile sprint collaborations with stakeholders, integrating feedback for continuous pipeline improvements and feature enhancements.

Analyzed complex AI-driven electricity demand data to generate actionable reporting insights and visualizations for Fortune 500 energy sector companies.

Designed and analyzed a student accessibility survey (n=34) using Qualtrics and Excel, identifying behavioral trends that informed adaptive program feasibility.

Wrote and optimized SQL queries to extract critical data from the ministry's housing database, supporting weekly updates on construction timelines and fund disbursement.

Monitored occupancy rates and housing demand trends, contributing to the reallocation of resources to high-need districts and improving unit utilization by 18%.

Developed and maintained interactive Tableau dashboards to visualize housing project progress across 15+ local government areas, reducing reporting time by 30%.

Conducted customer surveys with 100+ respondents, generating insights that increased customer engagement by 15%.

Analyzed 60+ competitors in Nigeria, producing detailed reports that improved market positioning and pricing strategies.

Enhanced crop health classification model precision by 20% through annotation of 10,000+ labeled images.

Supported pilot deployment of machine learning tools to 500+ farmers, enabling data-driven crop management decisions.