Build a fault-tolerant Monitoring Pipeline
Last updated: October 28, 2025
Quick Overview
Design a fault-tolerant monitoring system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
9
4
3,616 solved
Design a fault-tolerant monitoring system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
LinkedIn asks this during the System Design Round to assess your understanding of the full ML lifecycle. They want to see how you translate a business problem into an ML objective, design the feature pipeline, and plan for model monitoring and retraining.
What the Interviewer Expects
- Define clear ML objectives with appropriate loss functions and metrics
- Design a comprehensive feature engineering pipeline
- Discuss model selection with trade-offs (complexity vs interpretability vs latency)
- Plan online and offline evaluation strategies including A/B testing
- Address serving infrastructure: batch vs real-time, latency requirements
- Consider data quality, labeling strategy, and feedback loops
Key Topics to Cover
How to Approach This
- Start by clarifying functional and non-functional requirements with the interviewer.
- Estimate the scale: QPS, storage, bandwidth. This drives your design decisions.
- Draw a high-level architecture first, then deep dive into 1-2 critical components.
- Discuss trade-offs explicitly (e.g., consistency vs availability, SQL vs NoSQL).
- Address failure scenarios, monitoring, and how the system handles 10x traffic spikes.
Possible Follow-up Questions
- What is your model retraining strategy?
- How would you handle a 10x increase in prediction requests?
- What would you do if model performance degrades over time?
- How would you run A/B tests on different model versions?
Practice a Similar Problem on Codemia
Solve a related problem with our interactive workspace, get AI feedback, and view detailed solutions.
Solve on CodemiaSample Answer
Requirements
Functional Requirements
- Data Ingestion: The system must handle millions of incoming monitoring requests per second from various LinkedIn services (e.g., user activity, job postings).
- **R...
Capacity Estimation
Assuming LinkedIn receives approximately 100 million user interactions daily:
- Requests per Second: 100 million interactions / 86,400 seconds = ~1,157 requests per second.
- Data Volume: Each...
Submit Your Answer
