← Stripe Interview Insights

Stripe·Machine Learning Engineer·Onsite - System Design / Architecture·Senior

Senior
May 2026

Summary

Stripe ML engineer system design round, one big open-ended question about fraud detection that basically ate the whole session. Dense topic and they clearly wanted depth on every layer, not just the model part.

Questions Asked (1)

Q1

Design a machine learning system for fraud detection in an online payment platform that scores transactions in real time and supports approve, block, challenge, or manual review decisions.

System DesignTechnical Trade-offsProduct Analytics & Metrics
Author's notes

This is a beast of a question.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the business context: transaction volume, latency requirements, fraud cost trade-offs, and existing infrastructure. Then design a layered system: real-time feature computation, a low-latency model for scoring, and a decision engine that maps scores to actions (approve, block, challenge, manual review). Finally, discuss how to handle feedback loops, model updates, and monitoring for drift and adversarial adaptation.

Pro tip: Emphasize the importance of separating the fraud score from the decision policy—the score is a model output, while the decision thresholds should be configurable and based on business costs. Also, mention that you'd start with a simple model (e.g., logistic regression) and iterate, rather than jumping to deep learning.

1. Clarify Requirements and Constraints

Ask about transaction volume, latency SLA (e.g., <100ms), acceptable false positive/negative rates, and integration with existing systems. Understand the cost of different errors and the need for explainability.

2. Design Data and Feature Pipeline

Outline real-time feature computation (e.g., velocity checks, user history, device fingerprinting) using a stream processing framework. Discuss feature store for consistency between training and serving.

3. Model Selection and Training

Choose a model balancing accuracy and latency (e.g., gradient boosted trees or logistic regression). Address class imbalance, concept drift, and the need for periodic retraining with labeled data.

4. Decision Engine and Actions

Map model scores to actions using thresholds optimized for business costs. Explain how to handle challenge (e.g., 3DS) and manual review queues, and how to incorporate rules for known fraud patterns.

5. Monitoring, Feedback, and Iteration

Set up monitoring for latency, score distribution, and fraud rate. Use feedback from manual reviews and chargebacks to retrain models. Discuss A/B testing for policy changes.

Key Points to Mention

  • Real-time feature engineering with low-latency constraints (e.g., using Kafka, Flink, or Redis).
  • Handling class imbalance and concept drift in fraud detection.
  • Trade-offs between model complexity, latency, and interpretability.
  • Decision policy optimization based on business costs (false positives vs. false negatives).
  • Feedback loops: using manual review outcomes and chargebacks as labels.
  • Monitoring for adversarial attacks and model degradation.

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