Design a Real-Time Fraud-Scoring Stream for a Card Issuer
Overview
What this challenge is about.
Design a Real-Time Fraud-Scoring Stream for a Card Issuer. Advanced challenge in design. Designing real products under real constraints, earn a blockchain-ve...
The Brief
What you'll do, and what you'll demonstrate.
Design a Kafka plus Flink (or Kafka Streams) pipeline that scores each card authorization for fraud within 80 ms at the p99 percentile while sustaining 600 events per second.
This is not a design exercise. It is the work a product designer does between a brief and a shipped interface. That distinction matters to every hiring manager who has seen candidates redesign Spotify's homepage and none who have worked under real product constraints.
When you finish, you will have something most graduates do not: a real-world deliverable, verified by Ewance, that you can show to a hiring manager and say "I did this. Here is the proof."
Earning criteria — what you'll demonstrate
- Design a stateful stream topology that maintains rolling per-key features and joins them against a live event stream
- Integrate a pre-trained scoring model into a streaming hot path within a strict latency budget
- Reason about event-time ordering, watermarks, and state recovery in a real-time pipeline
- Benchmark a streaming system at sustained and burst load and interpret tail-latency percentiles
- Catalog operational failure modes and design graceful degradation for a default-approve system
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Apache Flink
Apply apache flink to solve real industry problems and demonstrate production-level capability.
- Kafka Streams
Apply kafka streams to solve real industry problems and demonstrate production-level capability.
- Stream Processing
Apply stream processing to solve real industry problems and demonstrate production-level capability.
- Low Latency Systems
Apply low latency systems to solve real industry problems and demonstrate production-level capability.
- State Stores
Apply state stores to solve real industry problems and demonstrate production-level capability.
- Feature Engineering
Apply feature engineering to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Streaming Backend Engineer
Builds the real-time data pipelines that power fraud, pricing, and personalization. This challenge mirrors that work end to end: designing a Kafka and Flink topology, maintaining stateful rolling features, and proving the system meets a hard latency budget under production-like load.
This challenge sharpens
- apache-flink
- stream-processing
- state-stores
Low-Latency Platform Engineer
Owns the hot paths where milliseconds decide revenue or risk. The challenge builds exactly this muscle — integrating model inference into a streaming pipeline, benchmarking tail latency at sustained and burst load, and tuning the design until p99 holds under budget.
This challenge sharpens
- low-latency-systems
- stream-processing
- feature-engineering
Fraud Platform Engineer
Engineers the decisioning systems that score transactions for risk in real time. This challenge bridges directly: joining live authorizations with customer history, applying a scoring model per event, and designing graceful default-approve behavior when components fail.
This challenge sharpens
- kafka-streams
- feature-engineering
- state-stores