Vodralek started in 2014 with one specific problem: existing courses on machine learning used toy datasets, while finance professionals needed practice with instruments they actually encountered — bonds, equities, structured products. The gap between academic ML and financial practice was wide, and it showed in how graduates performed.
The curriculum grew from that observation. Each module addresses a concrete classification task: distinguishing fixed-income from equity instruments by their return profiles, identifying momentum signals in price series, or labelling volatility regimes using unsupervised clustering. Lessons use publicly available market data so learners can reproduce results independently.
