← PayPal Interview Insights

PayPal·Data Scientist·Onsite - System Design / Architecture·Senior

SeniorPrefer not to say
May 2026

Summary

PayPal data scientist interview that was basically one massive system design question about fraud detection. The scope was huge and I'm not sure I covered everything they wanted, but it was a genuinely interesting problem to work through.

Questions Asked (1)

Q1

You're given a large multi-table dataset of transactions and customer/merchant metadata with delayed or partial fraud labels. Design an end-to-end system to detect fraudulent transactions, covering feature engineering, class imbalance, label latency, training/validation splits, thresholding for review capacity, real-time scoring, feedback loops, monitoring, and backtesting.

System DesignTechnical Trade-offsData Modeling
Author's notes

This was the whole interview, basically.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Structure your answer around the full ML lifecycle, emphasizing how you handle delayed labels and class imbalance while aligning with business constraints like review capacity. Start with data understanding and feature engineering, then move through training, deployment, and monitoring, explicitly addressing each component in the question. Use concrete examples and trade-offs to demonstrate practical experience.

Pro tip: Quantify the impact of your design choices—e.g., how thresholding based on review capacity directly affects precision/recall and operational cost—and mention how you'd validate with backtesting using time-based splits to avoid leakage.

1. Data Preparation & Feature Engineering

Explore the multi-table dataset, join transactions with customer/merchant metadata, and engineer features that capture fraud patterns (e.g., velocity, deviation from norms, graph-based features). Handle missing values and ensure features are available at scoring time.

2. Handling Label Latency & Class Imbalance

Address delayed/partial labels by using techniques like PU learning, label propagation, or time-windowed labeling. Mitigate class imbalance with resampling, class weights, or anomaly detection methods, and evaluate using metrics robust to imbalance (e.g., PR-AUC).

3. Model Training & Validation Strategy

Use time-based splits to mimic real-world deployment and avoid leakage. Train models (e.g., gradient boosting, neural networks) with proper validation, and calibrate probabilities. Consider ensemble or online learning for adaptability.

4. Thresholding & Real-Time Scoring

Set thresholds based on review capacity and business costs (e.g., cost-sensitive learning). Deploy a low-latency scoring service with feature store integration, and implement a tiered review system (auto-approve, manual review, auto-block).

5. Feedback Loops, Monitoring & Backtesting

Incorporate analyst feedback to update labels and retrain models. Monitor data drift, model performance, and business metrics. Backtest on historical data with time-based splits to simulate production and estimate impact.

Key Points to Mention

  • Time-based validation to prevent data leakage and simulate real-world deployment.
  • Cost-sensitive thresholding aligned with review capacity and business KPIs.
  • Handling delayed labels via PU learning, label propagation, or time-windowed labeling.
  • Feature engineering with velocity, aggregation, and graph-based features.
  • Real-time scoring architecture with low-latency feature serving and model inference.
  • Monitoring for data drift, model degradation, and feedback loop integration.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.