AI & Machine Learning Application

• •Supervised-learning problem formulation
• •Classification and regression reasoning
• •Model-selection reasoning
• •Feature selection and feature-engineering considerations
• •Training/validation separation
• •Overfitting recognition
• •Bias–variance reasoning
• •Model comparison
• •Hyperparameter reasoning
• •Appropriate use of baseline models

Data & Analytical Reasoning

• •Data-quality assessment
• •Missing-data reasoning
• •Outlier investigation
• •Dataset segmentation
• •Exploratory data analysis
• •Problem decomposition
• •Hypothesis formation
• •Evidence-based investigation
• •Identification of potentially misleading variables

Model Evaluation & Diagnostic Reasoning

• •Precision
• •Recall
• •F1
• •Class imbalance
• •False-positive / false-negative consequences
• •Validation strategy
• •Error patterns
• •Model generalization
• •Threshold implications
• •Appropriate metric selection

Depth & Maturity of Understanding

• •Causal understanding
• •Conditional understanding
• •Trade-off reasoning
• •Boundary awareness

Learning & Adaptability Evidence

• •Rapid comprehension of changed constraints
• •Appropriate revision of an initial hypothesis
• •Transfer of an underlying modelling principle to a different scenario
• •Willingness to identify uncertainty rather than invent certainty
• •Ability to explain why an earlier assumption needed correction