I lead the Data & AI function at BTA, managing multiple squads of data scientists and engineers building probabilistic models, data platforms, and AI products for sports betting and racing markets.

My background is in statistical inference and probabilistic modelling. I completed a PhD at the University of Melbourne under Prof. David Balding and Dr. Yao-ban Chan, developing novel Markov chain Monte Carlo methods for sampling ancestral recombination graphs from DNA sequence data. The work was published in PLOS Computational Biology and released as an open-source Python package (ARGinfer). I also spent time as a visiting researcher at the Big Data Institute, University of Oxford.

• Sports prediction & pricing — Probabilistic models for Same Game Multi (SGM) and player prop markets across NBA, Soccer, Tennis, Rugby, and UFC
• Horse racing — Speed modelling, finishing time prediction, race simulation, and settling position models across 300K+ races
• Customer data science — Customer Lifetime Value estimation, VIP identification, promo/bonus abuse detection, syndicate behaviour analysis
• Data platform — End-to-end architecture on Azure and Databricks with Medallion design, Service Bus pipelines, and Key Vault integrations
• LLM applications — AI-powered Racing Chatbot integrating structured data with large language models

• Mahmoudi A., Koskela J., Kelleher J., Chan Y., Balding D. — Bayesian inference of ancestral recombination graphs, PLOS Computational Biology, 2021
• Mahmoudi A., Balding D., Chan Y. — A Probabilistic Model to Infer the Ancestral Recombination Graph, 30th International Biometrics Conference, 2020
• Mahmoudi A., Balding D., Chan Y. — Inference Under the Exact Coalescent with Recombination, Probabilistic Modelling in Genomics Conference, France, 2019

• ARGinfer — MCMC inference on ancestral recombination graphs from DNA sequence data