Strengthened students’ foundations in logic, set theory, and graph theory through interactive workshops and personalized support on assignments and exams.

Under Pr. Omer Nguena Timo Landry (Ph.D), conducted a comparative evaluation of ML architectures (MLPs, LSTMs, GANs, Transformers/LLMs) to identify ideal use-case scenarios based on data type, memory needs, and expected output behavior.
Applied learnings directly to capstone project by selecting an appropriate MLP architecture for tabular prediction tasks.

Led regulatory compliance processes (validated 50+ ingredients, drafted 100+ technical and safety data sheets, reviewed marketing claims), reducing approval time by 30% (from 14 to 10 days) while ensuring 100% compliance.
Managed the regulatory database (updated 60+ product records) and coordinated 15+ hygiene and safety audits across the Île-de-France network with no major non-compliance findings.

Built a reproducible pipeline to clean and merge multisource lab datasets (GC, HPLC-UV, extraction assays) using Excel workflows including pivot models, lookup systems and structured tables.
Identified variables influencing yield and antioxidant content through regression analysis, ANOVA, correlation matrices, and outlier detection.
Delivered actionable insights to industrial partners (Morocco, Turkey, India), including optimized extraction conditions that improved yield by 4.5%.

Built a MATLAB predictive classification model to optimize gold nanoparticle synthesis for encapsulated drug delivery, using a 1500+ lab samples ETL data pipeline to identify key reaction parameters impacting different properties (magnetic and plasmonic).
Achieved 54%+ prediction accuracy and was offered a Master of Research position from the PHENIX research team.