• Lead the design and development of ML systems that solve complex, ambiguous business problems • Make sound technical decisions on model architecture, evaluation methodology, and tradeoffs • Set standards for model validation, testing, and monitoring across the team • Identify when "good enough" is appropriate vs. when deeper investment is warranted • Debug and troubleshoot models that fail in production - understand why they fail, not just that they fail • Frame business problems as well-defined ML tasks with clear success criteria • Build robust predictive models (classification, regression, time series, causal inference) • Implement rigorous train/validation/test methodology to ensure real-world generalization • Identify and prevent data leakage, overfitting, and other failure modes before they reach production • Define metrics that align model performance with actual business outcomes • Conduct holdout testing on true out-of-sample data - recognize when CV metrics are misleading • Design and analyze experiments to measure causal impact • Communicate model limitations, uncertainty, and risk to technical and non-technical stakeholders • Partner with product, engineering, and business teams to ensure ML solutions solve real problems • Translate complex technical concepts into actionable recommendations for stakeholders • Contribute to hiring and technical interviews