Études de cas
Réalisations
Études de cas anonymisées.
AI-guided recommendation engine for premium floral e-commerce
A production-oriented recommendation system that guides customers through emotionally loaded floral purchases — using a deterministic state machine with LLM components constrained to intent parsing and rationale generation only.
RésultatGuided purchase flow in production: auditable recommendations, business rules editable without a deploy
Python · FastAPI · LangGraph · OpenAI · +6
CLV-driven acquisition & budget allocation
Predictive CLV modeling combined with lookalike audience optimization to shift acquisition from volume metrics to value-based targeting.
RésultatAcquisition budget targeted on predicted subscriber value instead of installs
Python · CLV Modeling · Meta Marketing API · Lookalike Audiences · +1
Call volume forecasting per operator
Bayesian hierarchical model for predicting call volumes at the operator level, enabling data-driven workforce planning and staffing decisions.
RésultatStaffing plan per operator with confidence intervals, replacing manual rules
Python · PyMC · Bayesian Inference · Poisson Regression · +2
Supplier risk intelligence framework
Network-based supplier risk modeling that identifies systemic vulnerabilities and "too-central-to-fail" suppliers before disruptions occur.
RésultatToo-central-to-fail suppliers identified before disruption, prioritised by systemic impact
Python · NetworkX · Graph Analytics · Risk Modeling · +2
Travel retail forecasting system
Multi-SKU demand forecasting pipeline for 30+ products across 50 duty-free stores with automated monthly updates.
RésultatStockouts ↓34%, inventory costs ↓18%; forecast error 38% → 24%
Python · Prophet · scikit-learn · PostgreSQL · +4
Consumer healthcare marketing mix model & budget optimizer
Proprietary Marketing Mix Model with budget optimization replacing intuitive allocation with data-driven decision making across multiple countries and touchpoints.
RésultatMedia budget set from marginal ROI across 10+ countries; marketing and finance on one number
Python · Bayesian Modeling · Optimization · Data Visualization · +2
B2B SaaS pricing optimization
Willingness-to-pay analysis and price sensitivity modeling for a B2B SaaS product, resulting in a new tiered pricing structure.
RésultatRevenue per customer ↑31%, enterprise conversion ↑45%
Python · pandas · scikit-learn · Qualtrics · +2
Vitamins product pricing optimization
Data-driven pricing strategy for new vitamin product launch using Bayesian demand modeling and mixed-integer programming to optimize portfolio-wide pricing.
RésultatRevenue ↑18%, portfolio margin ↑12% on launch pricing
Python · PyMC3 · PuLP · pandas · +3
Automated credit decisioning system
Machine learning-powered credit decisioning system leveraging alternative data from social media and digital footprints combined with traditional credit data to improve risk assessment accuracy and speed.
RésultatDecisions 10x faster, operating cost ↓60%, default prediction +25%
Python · Pandas · Scikit-learn · XGBoost · +4
Noms des clients anonymisés. Projets décrits à haut niveau pour préserver la confidentialité.