About the Instructor
Josh Starmer
PhD in quantitative genetics. 1.8M+ YouTube subscribers. Renowned for making complex statistical concepts accessible through visual explanations and real-world analogies.
First Principles
Every ML decision comes down to probability and uncertainty. Decision trees split on features. Random forests average many trees to reduce overfitting. Bayesian inference updates beliefs with evidence.
AI Ecosystem Fit
Layer 4 (Models) — Interpretable AI. While deep learning dominates headlines, tree-based and Bayesian models are more explainable and often outperform neural nets on structured data (80% of real-world data).
Prerequisites
What Paid Projects This Unlocks
Build data-driven agents: decision trees, ensemble agents, Bayesian inference systems. Deploy interpretable AI agents.
3 Client Projects You Can Get Paid For
Real businesses have this problem right now. Build it, pitch it, charge for it.
Stack This With → (Unlocks Higher-Paying Projects)
Apply to real SMB projects (demand forecasting, churn prediction)
Understand when to use trees vs neural nets
Deploy interpretable models with AWS
Why This Matters
Most SMB data is tabular (spreadsheets), not images/text. Tree-based models are often the best fit. Auditors demand explainability.