About Projects Skills Experience Contact
All Work

Thirteen projects,
thirteen industries

← Back to portfolio
Selected Work
13
03 ML · Railways
SNCF : Railway Incident Prediction

Analyzed weather-to-incident correlation across SNCF's network. Engineered features (is_rush_hour, is_weekend, temperature delta, wind speed, season flags) and built a predictive model for daily default counts to enable proactive maintenance scheduling.

Python scikit-learn Feature Engineering Time-Series
↗ R² = 0.71 · RMSE = 130 defects/day
View project
04 Graph Analysis · Gaming
Asmodee / BGA : Player Analytics & Game Network

Ingested Board Game Arena's 1.92M player dataset, compressed to Parquet and queried with DuckDB for fast analytics. Segmented players into Casual / Regular / Hardcore clusters, built a bipartite player↔game graph and computed Jaccard similarity coefficients to map game affinity across the catalog.

DuckDB Parquet SQL Graph Analysis Clustering
↗ 1.92M players segmented · Jaccard game similarity map
View project
05 NLP · Defense
Ministère des Armées : AI Text Detection

Built a binary classifier for the French Ministry of Defense to detect AI-generated text vs. human-written content. Trained on a custom dataset with NLP preprocessing and fine-tuned classification architecture.

NLP Classification LLMs Python
↗ 98% detection accuracy
View project
06 Analytics · Banking
BNP Paribas : Ops Dashboard & SLA Modeling

Analyzed BNP's internal ticketing system to identify SLA bottlenecks. Applied quantile regression to benchmark fair resolution times per category. Designed a smart task-reassignment engine to route overloaded teams to underutilized ones.

Quantile Regression Dashboard Process Optimization Python
↗ BAU & Tax = 50% of total inefficiency identified
View project
07 Sales Optimization · FMCG
Henkel : Sales Forecasting & Budget Reallocation

Analyzed Henkel's B2B hardware distribution revenue by store and region. Built geographic heatmaps of top-performing departments, modeled advertising ROI correlation, and produced concrete reallocation recommendations for commercial managers across Leroy Merlin and BricoMarché networks.

Sales Forecasting Geo Segmentation Ad ROI Modeling Data Viz
↗ High-potential store clusters identified for budget reallocation
View project
08 Optimization · Retail
Carrefour : SKU Assortment Optimization

Modeled optimal product mix across Conserves, Bières and Petfood categories under hard store constraints (min. 5-store regional rollout before listing; max. 20% reference movement per period). Quantified cannibalization and substitution rates per SKU to maximize category revenue.

SQL Optimization Cannibalization Modeling Retail Analytics
↗ 3 categories optimized under multi-constraint SQL model
View project
09 Data Quality · Insurance
Generali : IT Data Audit & LLM Agent Design

Audited 15,000+ rows of IT project management data: uncovered 640+ duplicates, 1,030 overlapping missions, 78% missing-data rate, and 31 contracts shorter than days worked. Designed GENERALI-A : an LLM-powered chatbot with Dynamic Query Reinforcement and per-project AI reliability scoring.

Data Cleaning LLM Agent SQL Power BI
↗ 40,000-line cleaned dataset · GENERALI-A agent designed
View project
10 Graph Theory · Maritime · Sustainability
CMA CGM : Maritime Route Optimization

Modélisé le réseau maritime de CMA CGM (500+ navires, 420+ ports) comme un graphe pour calculer des routes optimales selon trois critères : rapidité, coût économique et émissions CO₂. Intégré un modèle météo/vent pour ajuster la vitesse réelle des navires, et proposé des solutions de décarbonation (kite Seawing, rotors Flettner).

Graph Theory Network Optimization Python Maritime Logistics CO₂ Modeling Sustainability
↗ −33% coût · −19k kg CO₂ vs route la plus rapide · 5 ports clés identifiés
View project
11 ML · Finance · Reinforcement Learning
FinoVera : AI Investment Advisor

Application d'investissement IA générant des recommandations de portefeuille personnalisées. Trois approches ML en parallèle : Random Forest/XGBoost pour la classification hausse/baisse, LSTM sur fenêtres 30 jours pour la prédiction de variation, et PPO (reinforcement learning) pour l'optimisation dynamique du portefeuille. Sentiment analysis VADER fusionné avec données Yahoo Finance et indicateurs macro (FRED).

Python LSTM PPO / RL Random Forest VADER NLP Streamlit yfinance Swift
↗ LSTM 58.4% accuracy · PPO 10–20k timesteps · 3 approches ML
View project
12 Strategy · Luxury
LVMH / Louis Vuitton : Digital Strategy

3-week Business Deep Dive on LV's digital commerce with sales, customer cluster and product data. Identified South Korea and Middle East as priority growth vectors; designed AI-backed personalization framework, ultra-HNW client targeting model and 2026 sales forecasting architecture.

Customer Segmentation Market Analysis AI Strategy Luxury
↗ 2–4% of clients = 65–80% of growth modeled
View project
13 Marketing Analytics · Media
Jellyfish : Media Budget Optimization

Data-driven media plan for a French outdoor equipment brand launching an eco-responsible product line. Allocated $1M across 5 ad channels (Meta, Snap, TikTok, DV360, Google Ads) using historical campaign data, conversion benchmarks and audience analysis.

Media Planning Marketing Analytics Budget Optimization Paid Advertising
↗ Google Ads: 11% → 37% reallocation based on conversion data
View project