I bring together process engineering, modelling, and machine learning to make water infrastructure smarter, cleaner, and more resilient.
Who I am, and how I approach the work.
I work at the intersection of water and wastewater engineering, mathematical modelling, data science, and process control. My focus is on developing intelligent monitoring, decision-support, and control frameworks that help make water infrastructure more resilient.
In practice, this means building mechanistic and data-driven models, uncertainty-aware soft sensors, forecasting tools, and control systems—from digital tools for engineering design to operational digital twins for full-scale facilities. My work spans applied research and digital product development, with an emphasis on turning complex engineering workflows into practical tools that engineers and operators can use.
This site is where I share the research, projects, and writing I am most proud of, along with how I think about problems at the intersection of engineering, artificial intelligence, and the ideas that shape both.
Process knowledge and machine learning are stronger together. Hybrid approaches capture what pure theory or pure data each miss on their own.
Real plants are noisy and data-scarce. I build soft sensors and models that quantify what they don't know, not just what they predict.
I don't follow the hype and believe that the best solution is the simplest possible. I also favour performance you can actually explain to an engineer.
While I fall in love of theory — my goal is a decision an engineer can act on. I care about the last mile from model to practice.
From municipal design offices to research labs to applied data science in industry.
Edmonton, AB, Canada
Québec, QC / Remote
United States / Remote
Edmonton, AB, Canada
Canada / Egypt / Saudi Arabia
Cairo, Egypt
Cairo, Egypt
University of Alberta · Edmonton, AB, Canada
Dissertation: mechanistic and machine-learning modelling of nitrous-oxide emissions from wastewater treatment. Supervisors: Prof. Yang Liu & Prof. Peter A. Vanrolleghem.
Ain Shams University · Cairo, Egypt
Thesis: a low-cost fabric-filter system for wastewater treatment in small communities.
Ain Shams University · Cairo, Egypt
↳Engineering know-how lived in spreadsheets and experts' heads — hard to reuse, validate, or scale.
Technical lead defining the computation logic, subject-matter-expert validation workflows, and implementation path for practical decision-support tools used across water-sector engineering.
↳Utilities need to know critical process states they can't afford to measure directly.
Developing hybrid mechanistic/ML and probabilistic models that estimate key water-quality states — and quantify their own uncertainty — for data-scarce treatment plants.
↳Nitrous oxide is a potent greenhouse gas from treatment plants, yet notoriously hard to predict.
Doctoral research building mechanistic and interpretable machine-learning models — with feature selection and uncertainty analysis — to predict and help mitigate N₂O emissions at full and pilot scale.
↳Membrane fouling quietly degrades RO performance and is costly to manage reactively.
Postdoctoral work developing deep-learning forecasting models and control-oriented data pipelines for RO membrane-fouling prediction and recovery optimization.
↳Water utilities react to failures instead of anticipating them.
Connecting asset condition, process reliability, hydraulic constraints, energy, cost, and risk into predictive-maintenance and operational-intelligence workflows.
Peer-reviewed research on modelling, machine learning, and emissions in water and wastewater systems.
Have an idea that you'd like to discuss, or a project you think I can help with?
Feel free to drop me a line!