About Me
I'm a Senior Data Scientist with 8+ years building production probabilistic AI systems across sports analytics, pharma, and tech. I specialize in Bayesian ML, causal inference, and time series forecasting, with a proven track record of translating complex statistical models into measurable business impact.
I co-founded PyMC Labs and helped scale it from $0 to $1M profit in 2 years, serving Fortune 500 clients in pharmaceutical R&D (RNA vaccine development), agtech (crop optimization), and e-commerce. As a core contributor to PyMC (8K+ GitHub stars), I've implemented novel distributions and authored tutorials that serve as official documentation, educating thousands of practitioners worldwide.
In 2019, I created the Learning Bayesian Statistics podcast, which has grown to 12,000 monthly listeners and ranks in the top 1.5% globally. Through my online education platform Intuitive Bayes, I've trained 200+ students in probabilistic AI, generating $100K+ in revenue.
Currently, I'm a Senior Applied Scientist at the Miami Marlins, where I build Bayesian forecasting models processing 50K+ tracking events per game to predict player performance and inform $100M+ roster decisions.
Career Highlights
Miami Marlins
PyMC Labs
Intuitive Bayes
Learning Bayesian Statistics
PyMC
pollsposition.com
Key Impact & Achievements
Technical Skills
Machine Learning & AI
Advanced Methods
Tools & Frameworks
Applied Methods
Domain Expertise
Leadership & Communication
Selected Projects
Soccer Factor Model (2024)
Hierarchical Bayesian model decomposing player performance through time, into skill vs. team effects. Introduced novel metrics (Skill & Performance Above Replacement) for fair cross-player comparisons. Built interactive dashboard visualizing 1000+ player rankings with uncertainty estimates.
PyMC Technical Documentation (2019-Present)
Authored 8+ comprehensive tutorials on advanced topics including Gaussian Processes (HSGP), Multilevel Modeling, Kronecker Structured Covariances, and LKJ Cholesky Priors. These tutorials have been adopted as official PyMC documentation and have educated thousands of practitioners worldwide.
Intuitive Bayes - Online Education Platform (2021-Present)
Developed comprehensive online education platform to demystify probabilistic AI and take beginners to practitioners quickly. Created concise video lectures integrated with Python code and PyMC implementations. Trained 200+ students and generated $100K+ in revenue.
Electoral Forecasting Platform - pollsposition.com (2017-Present)
Built first French polling aggregation website using Multilevel Regression with Post-Stratification (MRP) models for French elections. Developed Gaussian Process and Hidden Markov Models to predict presidential approval ratings. Entirely open-source implementation with 4K visitors/month during electoral campaigns.
Publications & Papers
Hidden Diversity of Threatened Sharks and Rays in the Global Meat Trade
Currently under review in Science
Co-author and principal technical modeler of complex biology model evaluating hidden diversity of threatened species in global meat trade. Developed advanced bespoke Bayesian hierarchical model to evaluate conservation biology research with significant environmental impact.
Unveiling True Talent: The Soccer Factor Model for Skill Evaluation
arXiv, 2024
Developed novel hierarchical Bayesian model decomposing player performance through time, to isolate player skill from team effects. Introduced two new metrics (Skill & Performance Above Replacement) enabling fair cross-player comparisons.
Technical Blog Posts & Tutorials
How popular is the President?
Experimenting with a Gaussian Process to model presidential popularity across time.
Popularity hide and seek
Estimate latent presidential popularity across time with a Markov chain.
Gaussian Processes: HSGP Advanced Usage
Comprehensive guide to Hilbert Space Gaussian Processes for efficient GP approximations in PyMC.
A Primer on Bayesian Methods for Multilevel Modeling
In-depth introduction to hierarchical models with practical PyMC implementations.
Selected Talks & Presentations
Upcoming & Recent
A Beginner's Guide to State Space Modeling
PyData Berlin 2025 (Upcoming)
Introduction to state space models for time series analysis using PyMC.
A Beginner's Guide to Variational Inference
PyData Virginia 2025
Practical introduction to variational inference methods for scalable Bayesian inference.
Mastering Gaussian Processes with PyMC
PyData NYC 2024
Comprehensive tutorial on implementing and interpreting Gaussian Process models.
Live Show at Stancon Oxford 2024
Stan Conference, Oxford 2024
Live recording of Learning Bayesian Statistics podcast with audience participation.
Live Show at Imperial College London
Imperial College London
Interactive discussion on Bayesian methods in research and practice.
Podcast Appearances
Learning Baseball Through Statistics
Stats + Stories Podcast
Discussion on applying statistical methods to baseball analytics and player evaluation.
Bayesian Methods and Applications
Super Data Science Podcast
Overview of Bayesian statistics applications across industries and use cases.
Media Interviews
- Advancing Bayesian Statistics and Sports Analytics
- Harnessing Bayesian Intelligence: Intuitive Bayes Redefines Predictive Analytics in Sports and Beyond
- Redefining Analytics: Alexandre Andorra's Bayesian Revolution Across Sports and Data Science
- Cómo la estadística y la gestión de datos serán cada vez más determinantes en el deporte (Spanish media)
Awards & Recognition
🏆 US Green Card EB1 - Alien with Extraordinary Abilities
Recognized by United States government for extraordinary ability in STEM field. The EB1 category is reserved for individuals who have risen to the very top of their field of endeavor.
🏆 Fellow Member of BCS, The Chartered Institute for IT
Fellowship recognizing outstanding achievement in the field of information technology and computer science. FBCS is the highest grade of membership.
🏆 Cutting-Edge Data Science Innovator
Recognition by Topmate for innovative contributions to data science and education.
📊 Top 1.5% Global Podcast
Learning Bayesian Statistics podcast ranks in top 1.5% of all podcasts globally (out of 3+ million shows), with 12,000 monthly listeners.