Hadrien Kanso

🧠 Are you losing money on poorly exploited data? I transform your industrial, sensor, or business data into predictive models in production, not POCs gathering dust.

Senior Data Scientist · PhD in Applied Mathematics · 7 years of experience between research (INRAE, Aix-Marseille) and industry (Expleo Group).

Measured results:

→ -40% energy consumption on industrial HVAC systems (machine learning + physics)

→ -35% unplanned downtime via predictive maintenance (Random Forest + neural network, F1-score 97%)

→ -50% complexity of dynamic models without loss of robustness (PhD INRAE)

→ Predictive wildfire models delivered to field teams (CEREN)

What I solve for you:

→ Predictive modeling & applied machine learning (Python, TensorFlow, PyTorch, Scikit-Learn)

→ Predictive maintenance & anomaly detection on time series

→ Energy optimization through AI hybrid physics + deep learning models

→ Industrialization: data pipelines, MLOps, monitoring, drift detection

→ Reduction of complex models, dimension reduction

→ Tailored training in data science, machine learning, deep learning, statistics, Power BI

Sectors: Energy · HVAC · Industry 4.0 · Fire safety · Greentech · Agri-food · R&D

Stack: Python · Machine Learning · Deep Learning · TensorFlow · PyTorch · Scikit-Learn · Pandas · NumPy · Time Series · SQL · Matlab · R · Power BI

My difference: I understand the physics of the problem before modeling. This is what separates a model that works in production from a POC that will never be deployed.

Founder of Kanova Labs AI for energy optimization and industrial decarbonization.

📩 Describe your problem to me — I'll respond in under an hour.