Esmaeel Mohammadi

I help energy, water, and industrial companies optimize complex physical systems using AI, digital twins, and mathematical modeling.

I am an engineer who speaks both the language of physics and the language of AI. While many data scientists focus on dashboards or NLP, my work is in Industrial AI — applying reinforcement learning, optimization, and hybrid physics-AI models to real-world infrastructure such as wastewater treatment plants, energy systems, and other large-scale physical processes.

My core focus includes:

Control & Optimization

Designing reinforcement-learning and MPC-based controllers that go beyond forecasting and actively optimize real-world processes.

Digital Twins

Building high-fidelity simulation environments from time-series data and physical models to test, validate, and deploy AI control systems safely.

Scientific Machine Learning

Combining classical engineering (ODEs, thermodynamics, mass balances) with modern deep learning to create models that are accurate, stable, and deployable.

I hold a PhD in Data Science & Engineering Cybernetics (Marie Skłodowska-Curie Fellow), where I developed deep-RL agents for the autonomous control of biological processes in wastewater treatment. I currently work in the energy sector, developing simulation environments and reinforcement-learning systems for energy procurement and consumption forecasting.

I typically support organizations with:

AI-driven process optimization

Digital twin development

Time-series forecasting & decision support

Feasibility studies, PoCs, and transition from research to production

I am a builder at heart: I take ownership of complex mathematical problems, implement solutions in Python, PyTorch, and C++, and deliver systems that can be used in production environments.

If you are working with energy systems, utilities, industrial plants, or complex physical assets and want to move from static models to AI-driven optimization, feel free to reach out.