Build a predictive model from TurboTax sample data
Company: Intuit
Role: Data Scientist
Category: Machine Learning
Difficulty: easy
Interview Round: Onsite
# Build a predictive model from TurboTax sample data
You receive a **TurboTax sample dataset** (user-level and/or session-level) and are asked to build a predictive model.
## Task
1. Pick a concrete prediction target (choose one and justify):
- Probability a user will **file** within 14 days of starting.
- Probability a user will **churn** (not file this season).
- Expected **revenue** from the user this season.
2. Describe how you would:
- Define labels and avoid label leakage.
- Build features from product interactions and historical attributes.
- Choose a baseline model and at least one stronger model.
- Evaluate performance (metrics, calibration, slice performance).
- Turn the model into an actionable recommendation (e.g., targeting, prioritization, interventions).
## Constraints / realism
- Data may be missing or delayed.
- Class imbalance is likely (e.g., churn).
- The business cares about interpretability and safe deployment.
### Constraints & Assumptions
- Preserve the scope, facts, inputs, and requested outputs from the prompt above.
- If the prompt leaves a detail unspecified, state a reasonable assumption before relying on it.
- Keep the answer interview-ready: concise enough to present, but concrete enough to implement or evaluate.
### Clarifying Questions to Ask
- Clarify the task, data shape, labels, constraints, and evaluation metric.
- State assumptions behind the math or modeling technique you choose.
- Connect theory to practical training, debugging, and deployment implications.
### What a Strong Answer Covers
- Correct definitions and formulas where the prompt requires them.
- A practical explanation of how the method behaves on real data.
- Trade-offs, failure modes, diagnostics, and mitigation strategies.
- Evaluation choices that match the product or modeling objective.
### Follow-up Questions
- How would noisy labels, class imbalance, or distribution shift affect the answer?
- What would you monitor after deployment?
- Which baseline would you compare against first?
Quick Answer: Build a predictive model from TurboTax sample data evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.