Data Scientist

Build the "Gold Standard": You will design, train, and validate the core machine learning models (Classification, Regression, Forecasting) that solve our most critical business problems.

Rapid Prototyping: You will use Python to build functional proofs-of-concept. You deliver the validated logic and mathematical framework that our Engineering team will scale.

Algorithmic Strategy: You decide we solve a problem. Is this a rules-based engine A simple regression Or a complex deep learning solution You own the trade-off between complexity and business value.

Signal Discovery: You will not just use the data we have; you will dictate the data we . You will design the feature requirements and tracking events necessary to capture improved user signals.

Metric Definition: You will move beyond vanity metrics. You will define the "North Star" KPIs (e.g., Retention, CLV, Engagement Quality) that align our algorithms with actual business health.

Hypothesis Generation: You will mine our data landscape to find user friction points and propose data-backed features to fix them.

Statistical Rigor: You will design the framework for A/B tests and product experiments, ensuring we distinguish between random noise and actionable trends.

Roadmap Influence: You will present data narratives to leadership that directly influence we build next.

Proficient in Python SQL (ML Native): You are fluent in the Data Science stack (Pandas, NumPy, Scikit-Learn) and complex SQL. You write clean code to explore data and build models, not to develop applications.

Strong Statistical Foundation: You have a deep grasp of probability, hypothesis testing, and error analysis. You understand a model works, not just how to import it.

Data Visualization: You can build compelling visualizations (Matplotlib, Seaborn) that make complex data obvious to non-technical stakeholders.

Outcome Over Output: You care more about moving a business metric than optimizing a hyperparameter. You understand that a simple model that solves a user problem is infinitely better than a complex one that never ships. You optimize for business KPIs (ROI, Retention), not just model performance metrics (AUC/F1).

User Empathy: You constantly ask, "How does this model improve the users life"