Bird
Raised Fist0

Feed Ranking System - Twitter, Instagram, Facebook at Scale

SYSTEM-DESIGNhardInterview freq: very_high
Choose your preparation mode6 modes available
LearnTraceDebugTrade-offChallenge
▶
Execution Trace
Step 1/5
Data Ingest

User interaction data (clicks, likes) is collected and stored.

🎯 Understanding data sources is crucial for feature engineering.
data = collect_user_data()
store_raw_data(data)
Feat Eng

Raw data is transformed into feature vectors, e.g., user embeddings.

🎯 Feature engineering impacts model performance significantly.
features = transform_data(data)
store_features(features)
Model Train

Model training reduces loss from 0.95 to 0.45 over 10 epochs.

🎯 Model convergence is key to effective ranking.
model = train_model(features)
save_model(model)
Serving Layer

API serves ranked feed to users with latency < 200ms.

🎯 Low latency is critical for user experience.
ranked_feed = serve_feed(user_id)
return ranked_feed
Monitoring

System logs show a 5% increase in engagement post-deployment.

🎯 Monitoring helps in assessing the impact of model updates.
metrics = monitor_system()
log_metrics(metrics)
📊
Diagram
Diagram
┌───────────────┐     ┌───────────────┐     ┌───────────────┐
│ Data Ingest   │ ──▶ │ Feat Eng      │ ──▶ │ Model Train   │
│ (User, Item)  │     │ (Features)    │     │ (Deep NN)     │
└───────────────┘     └───────────────┘     └───────────────┘
       │                     │                     │
       ▼                     ▼                     ▼
┌───────────────┐     ┌───────────────┐     ┌───────────────┐
│ Data Storage  │ ◀── │ Feat Store    │ ◀── │ Model Store   │
│ (Raw, Feat)   │     │ (Transformed) │     │ (Weights)     │
└───────────────┘     └───────────────┘     └───────────────┘
       │                     │                     │
       ▼                     ▼                     ▼
┌───────────────┐     ┌───────────────┐     ┌───────────────┐
│ Serving Layer │ ◀── │ Ranker       │ ◀── │ Monitoring    │
│ (API)         │     │ (Predictions)│     │ (Metrics)     │
└───────────────┘     └───────────────┘     └───────────────┘
Labels
  • Data Ingest: Collects user and item data
  • Feat Eng: Transforms raw data into features
  • Model Train: Trains a deep neural network
  • Serving Layer: Provides ranked feed via API
  • Monitoring: Tracks performance metrics
🔑
Key Moments
▲
Q: Why is feature engineering crucial in this pipeline?
A: Feature engineering transforms raw data into meaningful inputs that significantly impact model performance and accuracy.
Q: How does the serving layer ensure low latency?
A: The serving layer is optimized to handle requests efficiently, ensuring that ranked feeds are delivered with minimal delay to enhance user experience.
Q: What role does monitoring play in the system?
A: Monitoring tracks system performance and user engagement, providing insights into the effectiveness of the model and guiding future improvements.
Trace Summary

The diagram illustrates a feed ranking system used by social media platforms like Twitter, Instagram, and Facebook. It shows the end-to-end ML pipeline from data ingestion, where user interactions are collected, to feature engineering, which transforms raw data into features. The model training phase uses these features to train a deep neural network. The serving layer provides ranked feeds to users via an API, ensuring low latency. Monitoring tracks system performance, allowing for continuous improvement. Understanding each component's role is crucial for designing scalable and efficient recommendation systems.

Primary Focus

✓ ML pipeline from data ingestion to serving

Practice

(1/5)
1. For a fine-tuning pipeline using RLHF, which metric is most appropriate to evaluate the alignment of model outputs with human preferences?
easy
A. Precision
B. Reward Score
C. Accuracy
D. F1 Score

Solution

  1. Step 1: Understand the purpose of RLHF

    RLHF aims to align model outputs with human preferences, which is best evaluated by a reward score reflecting human feedback.

  2. Step 2: Eliminate wrong options

    A: Accuracy is too generic. B: Precision is for binary classification. D: F1 Score is for balance between precision and recall, not alignment.

  3. Final Answer:

    Option B -> Option B

  4. Quick Check:

    Reward Score reflects human feedback. [OK]
Hint: Reward Score reflects human feedback. [OK]
Common Mistakes:
2. In the context of a Guardrails System for LLMs, which component is primarily responsible for detecting inappropriate content in user-generated text?
easy
A. Data augmentation module
B. Content filtering module
C. Model training module
D. User feedback loop

Solution

  1. Step 1: Identify the role of each component

    The content filtering module is designed to detect and filter inappropriate content.

  2. Step 2: Eliminate wrong options

    Data augmentation is for enhancing training data, model training is for learning patterns, and user feedback loop is for improving system iteratively.

  3. Final Answer:

    Option B -> Option B

  4. Quick Check:

    Content filtering detects inappropriate content. [OK]
Hint: Content filtering detects inappropriate content. [OK]
Common Mistakes:
3. In a search ranking system, which component is primarily responsible for determining the relevance of a document to a query?
easy
A. Query parser
B. Ranking algorithm
C. Indexing module
D. Crawler

Solution

  1. Step 1: Identify the role of each component

    The ranking algorithm is responsible for determining the relevance of documents to a query.

  2. Step 2: Eliminate wrong options

    Indexing module organizes data, query parser interprets queries, and crawler collects data.

  3. Final Answer:

    Option B -> Option B

  4. Quick Check:

    Ranking algorithm determines relevance. [OK]
Hint: Ranking algorithm determines relevance. [OK]
Common Mistakes:
4. As the user base scales 10x, which component of a DLRM-based CTR prediction system is most likely to become a bottleneck?
hard
A. Data preprocessing pipeline
B. Training data storage capacity
C. Model inference latency
D. Embedding table size

Solution

  1. Step 1: Consider scaling effects on components.

    Embedding tables grow with user and feature diversity.

  2. Step 2: Eliminate wrong options.

    Preprocessing (A) and inference (B) scale with compute resources. Storage (D) is less critical than real-time constraints.

  3. Final Answer:

    Option D -> Option D

  4. Quick Check:

    Embedding tables grow with scale. [OK]
Hint: Embedding tables grow with scale. [OK]
Common Mistakes:
5. As the number of LLMs evaluated increases tenfold, which component of the evaluation system is most likely to fail first?
hard
A. Human evaluator availability
B. Benchmark task diversity
C. Data storage capacity
D. LLM-as-Judge processing power

Solution

  1. Step 1: Analyze scalability of components

    Human evaluators are a bottleneck as scaling requires more human resources.

  2. Step 2: Eliminate wrong options

    Data storage and processing power can be scaled with infrastructure, and task diversity is not a direct scalability issue.

  3. Final Answer:

    Option A -> Option A

  4. Quick Check:

    Human evaluators are the bottleneck in scaling. [OK]
Hint: Human evaluators are the bottleneck in scaling. [OK]
Common Mistakes:
Recommended

Ready to prove it?

Pass the test and this pattern moves unknown strong while earning +1 Grip toward your certificate.

10 questions~5 mineasy hard