Book companion
Machine Learning labs
Scan a chapter QR or open a lab directly. You need your Adaptly account to open them.
Reader access
Open the book labs
Labs are for students with an account. Sign in or create your account to open them.
01Operators vs ArchitectsSystem thinking for machine learning02Competitions vs BusinessMoving from leaderboard thinking to product outcomes03Feature EngineeringTurning raw data into useful model signal04Data Is GeometryUnderstanding datasets through structure and distance05Correlation, Causation, and Why Your Model LiesLeakage, bias, and misleading patterns06The Interpretability ProblemExplaining model behaviour without pretending certainty07Evaluation as a DisciplineEvaluation beyond one headline metric08Building Systems, Not DemosProduction readiness and operational thinking09The Modern ML StackChoosing tools by system responsibility10RAG Done CorrectlyRetrieval as an evaluated system11Agent Architectures: Hype vs RealitySafe agent loops and control boundaries12Fine-Tuning vs PromptingCost, quality, and operational trade-offs13Quantization and Efficient InferenceLatency, throughput, cost, and quality14Building Your ML MachineLocal, cloud, and hybrid compute choices15API Dependence vs Computational SovereigntyProvider risk, routing, and auditability16The Math You Actually NeedPractical mathematical reasoning for ML engineers17The Data Science LayerData audit before model ambition18Competitive Programming and ML ThinkingAlgorithmic discipline inside ML systems19How Big Tech Evaluates EngineersStructured ML system design communication20What You Build NextTurning the book into a concrete project path
