Kheireddin KADRI
Doctor in soft matter · Data Science & numerical simulation
Researcher and trainer, Kheireddin bridges the rigor of scientific research with the industrial practice of AI: from a PhD in materials physics to deploying models in production.
Kheireddin KADRI – From fundamental research to applied AI
Kheireddin KADRI holds a PhD in soft matter and specializes in data science and numerical simulation. A postdoctoral researcher at ESILV since 2022, he defended his thesis at Arts et Métiers ParisTech on the role of shear in the rupture of ultra-thin polymer films. He now puts this scientific rigor at the service of artificial intelligence, data, and training.
A PhD at Arts et Métiers ParisTech
During his thesis (2017–2020), Kheireddin combined numerical and experimental studies to understand the rupture of ultra-thin polymer films under shear. He developed a method to solve the thin-film equation accounting for capillary and van der Waals forces, revealing two rupture regimes whose transition resembles a phase transition — a modeling rigor that still shapes his data practice today.
Postdoctoral researcher at ESILV
Since 2022, Kheireddin has pursued his research at the École Supérieure d'Ingénieurs Léonard de Vinci, at the crossroads of algorithms, AI, and systems modeling. There he supervises work blending deep learning and data science, from graph-based gait recognition (ST-GCN) to structure generation with generative adversarial networks.
Data science & machine learning
Trained in data science (Ironhack bootcamp, M2 in statistical modeling at UPMC) and certified Machine Learning Researcher (Workera) and in GANs (DeepLearning.AI), Kheireddin masters the full chain: data exploration, modeling, experiment tracking with MLflow, and production deployment.
MLOps & Microsoft Azure
Kheireddin teaches MLOps and prepares his students for the Microsoft Azure DP-100 certification (Azure ML Engineer Associate). ML pipelines, model deployment and monitoring, observability with Prometheus and Grafana, Azure AI / Foundry environments: he teaches a complete lifecycle, from training to production.
Generative AI & applied research
Passionate about generative AI, Kheireddin explores GANs (StyleGAN, MolGAN, generation of superconductor crystal structures), graph neural networks, and LLMs applied to materials science — notably at the KIT AIMat Summer School in Karlsruhe. He has led several workshops, including one at Ifremer Brest.
“Data does not replace understanding: it extends it. A good model always starts with a good question.”
Scientific rigor
every model is validated like an experiment: clear hypotheses, metrics, reproducibility.
From theory to practice
moving from concept to concrete implementation, all the way to production deployment.
Knowledge sharing
mentoring, teaching, and guiding learners toward technical autonomy.
Interdisciplinary curiosity
connecting physics, data, and AI to open up new use cases.
- MLOps & model lifecycle
- Microsoft Azure ML & DP-100
- Data science & machine learning
- Generative AI (GANs) & graph neural networks
- Deployment, monitoring (Prometheus, Grafana) & MLflow
- Numerical simulation & scientific modeling
Let's build
your next
program.
30 minutes with a learning advisor. No commitment. No sales pitch dressed up as a demo.