At Olam International, I built the ML and optimization backend for a global cocoa procurement web application, including yield prediction models and a linear programming optimizer. The work delivered USD 2M in annual savings through procurement optimization.
I also used Elastic Net regression to optimize freeze-dried coffee plant processes and developed an ARIMAX model for Ghana Cedi to USD forecasting. I led a team of three to productionize decision systems.
As a Data Science Consultant for a US-based Renewable Energy Company, I built an XGBoost pipeline to predict biogas output and deployed models on Azure ML as REST APIs. I also adapted a production-grade model to a newly commissioned plant using booster-level transfer learning and developed a feedstock optimization engine with SciPy.
Earlier at Olam International, I created Python optimization systems for cotton allocation and global shipping routes, contributing to annual savings. At LatentView Analytics, I built automated analytics systems and led a Tableau migration for an e-commerce client.
