I am Achraf Bouhmady, a postdoctoral researcher in applied mathematics and data science, with interests in optimization, uncertainty, and real-world decision systems, with a particular focus on optimal control, Hamilton-Jacobi-Bellman equations, viability theory, and quantitative finance.
As a Ph.D. candidate at Mohammed V University in Rabat, my research focused on control theory and machine learning, with work spanning robust optimal control, viability theory, Hamilton-Jacobi-Bellman equations, and PDE control. By integrating theoretical insights from control theory with data-driven approaches from machine learning, I aimed to contribute to efficient and robust algorithms for economics, engineering, and resource management.
This MIT program strengthened my ability to make informed decisions using data-driven approaches. Through hands-on projects, case studies, and practical applications, I developed experience applying machine learning and analytics techniques to real-world problems across different domains.
This program provided a strong foundation in algebra, functional analysis, mathematical modeling, optimization, and statistical analysis. It developed my analytical and problem-solving skills while preparing me to apply mathematical methods to finance, logistics, operations research, and data analysis.
Leading research on probabilistic predictive maintenance, with a focus on decision-making under uncertainty, risk-aware modeling, and long-term reliability in real-world systems.
Conducted advanced research in optimal control and developed numerical techniques for solving nonlinear problems related to the Hamilton-Jacobi-Bellman equation. Collaborated on interdisciplinary projects, published scientific articles, presented results at international conferences, and participated in seminars and workshops on optimal control and numerical methods.
Taught descriptive statistics, sampling methods, and hypothesis testing to Bachelor's degree students in economics and management. The goal was to help students collect unbiased data, analyze it correctly, and make informed decisions based on reliable statistical evidence.
Taught Mathematical Analysis and Numerical Analysis to students in mathematics and computer science. Covered functions, sequences, limits, integrals, and numerical methods such as Gauss and Cholesky, with practical implementation in MATLAB for solving nonlinear problems.
A data science portfolio that gets cited by AI should answer one concrete question fast, include verifiable numbers with sources, and show a reproducible workflow from data to decision.
The food aggregator company has stored order data from registered customers to understand restaurant demand and improve customer experience.
Comprehensive analysis of pricing and advertising strategies, market equilibrium, and strategic interactions in competitive environments.
A dynamic modeling framework combining game theory and optimization to analyze interacting agents in financial markets.
Customer prediction case in EdTech using data analysis to identify high-potential conversion profiles.
Nonlinear parameter estimation in MATLAB for models where relationships cannot be represented by simple linear equations.