← Snapchat Interview Insights

Snapchat·Machine Learning Engineer·Technical Phone Screen·Senior

SeniorPrefer not to say
Jun 2026

Summary

Applied ML interview at Snapchat for an MLE role, centered almost entirely on designing an ads ranking system from scratch. Pretty deep dive, they wanted the full stack from problem framing through serving constraints, not just a surface-level answer.

Questions Asked (6)

Q1

How would you design an ads ranking model end to end?

System DesignTechnical Trade-offs
Author's notes

This was the whole interview basically.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the business objectives and constraints, then walk through the ML lifecycle from data collection to deployment, emphasizing trade-offs at each stage. Focus on how you would handle Snapchat's unique aspects like ephemeral content and social context.

Pro tip: Demonstrate awareness of the feedback loop between ranking and user behavior, and how to mitigate biases like position bias. Also, discuss how you would measure long-term user satisfaction beyond immediate CTR.

1. Clarify Requirements and Metrics

Ask about business goals (e.g., revenue, user engagement), constraints (latency, scale), and define success metrics (e.g., CTR, conversion, long-term value).

2. Data Collection and Feature Engineering

Identify data sources (user interactions, ad content, context), handle missing data, and engineer features (user demographics, ad quality, context embeddings).

3. Model Selection and Training

Choose model architecture (e.g., deep learning for large-scale, gradient boosted trees for interpretability), define loss function (e.g., log loss, ranking loss), and train with proper validation.

4. Evaluation and Offline Testing

Use offline metrics (AUC, NDCG) and simulate online performance with counterfactual methods; validate with A/B tests.

5. Deployment and Monitoring

Deploy model with low-latency serving, set up monitoring for performance drift, and iterate with online learning if needed.

Key Points to Mention

  • Handling position bias and feedback loops in ranking
  • Trade-offs between model complexity and latency
  • Use of multi-task learning to optimize for multiple objectives
  • Incorporating social context and user-ad interaction features
  • Online evaluation with A/B testing and interleaving
  • Scalability and real-time inference considerations

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

Q2

How do you define labels for an ads ranking model, and what are the tradeoffs between using clicks versus conversions?

Data ModelingTechnical Trade-offs
Author's notes

CTR labels are noisier but abundant, CVR labels are sparse and delayed.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining the purpose of labels in ads ranking: they are the target variable the model learns to predict, and their choice directly impacts optimization. Then compare clicks vs. conversions in terms of signal availability, latency, and alignment with business goals, highlighting tradeoffs and potential solutions like multi-task learning.

Pro tip: Emphasize that the choice of label should align with the business objective and that using clicks as a proxy for conversions can introduce bias, especially if there's a discrepancy between user intent and advertiser value. Mention that at Snapchat, where visual and ephemeral content is key, engagement signals like clicks might be more immediate but conversions drive revenue.

1. Define the label and its role

Explain that labels are the ground truth events the model predicts, such as click or conversion, and they determine the model's optimization target.

2. Discuss clicks as labels

Highlight pros: abundant data, immediate feedback, and strong correlation with engagement. Cons: may not directly reflect advertiser value, susceptible to clickbait, and can be noisy.

3. Discuss conversions as labels

Highlight pros: directly tied to advertiser ROI and business goals. Cons: sparse data, delayed feedback, and attribution challenges.

4. Compare tradeoffs

Contrast the two in terms of data volume, latency, bias, and alignment with business objectives. Mention that clicks are a proxy but can lead to suboptimal long-term outcomes.

5. Propose solutions

Suggest approaches like multi-task learning (predicting both clicks and conversions), using clicks as auxiliary labels, or employing techniques like delayed feedback modeling to handle conversion latency.

Key Points to Mention

  • Label definition: the target variable for supervised learning in ads ranking.
  • Clicks: high volume, immediate, but may not align with advertiser value.
  • Conversions: sparse, delayed, but directly tied to ROI.
  • Tradeoffs: data sparsity, latency, bias, and business alignment.
  • Multi-task learning: jointly model clicks and conversions to leverage both signals.
  • Attribution and delayed feedback: challenges with conversion labels and potential solutions.

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

Q3

What features would you use in an ads ranking model?

Data ModelingSystem Design
Author's notes

Went through user features, ad features, context, and cross features.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the objective of the ads ranking model (e.g., maximize revenue, CTR, or conversions) and the constraints (latency, scale). Then, structure your answer around feature categories: user, ad, context, and interaction features, and explain how each contributes to ranking. Finally, discuss feature engineering techniques and evaluation metrics to show end-to-end thinking.

Pro tip: Emphasize the importance of feature freshness and real-time features for ads ranking, as Snapchat's ephemeral content demands up-to-date signals. Also, mention how you would handle missing features and monitor feature drift in production.

1. Clarify Objective and Constraints

Ask clarifying questions to understand the goal (e.g., CTR, CVR, revenue) and system constraints (latency, scale, privacy). This ensures your feature choices align with business needs.

2. Categorize Features

Organize features into logical groups: user features, ad features, contextual features, and interaction features. This makes your answer structured and comprehensive.

3. Detail Key Features per Category

Provide specific examples for each category, such as user demographics and past engagement for user features, ad creative and targeting for ad features, time and device for contextual, and user-ad similarity for interaction.

4. Discuss Feature Engineering and Selection

Explain how you would transform raw data into features (e.g., embeddings, counts, ratios) and select the most predictive ones using techniques like regularization or feature importance.

5. Address Production Considerations

Mention challenges like real-time feature computation, feature stores, handling missing values, and monitoring for drift. This shows awareness of operational aspects.

Key Points to Mention

  • User features: demographics, historical engagement (CTR, CVR), interests, and embeddings from user behavior sequences.
  • Ad features: creative attributes (image, text, video), advertiser quality, bid price, and targeting criteria.
  • Contextual features: time of day, day of week, device type, OS, network connectivity, and location.
  • Interaction features: similarity between user and ad embeddings, user-ad historical interactions, and cross features.
  • Real-time features: session-based features (e.g., recent views, swipes) and real-time user-ad affinity scores.
  • Evaluation metrics: offline metrics like AUC, log loss, and online metrics like CTR, CVR, and revenue lift.

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

Q4

How do you handle position bias in training data for an ads ranking system?

Data ModelingTechnical Trade-offs
Author's notes

Blanked for a second.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining position bias in ads ranking and its impact on training data, then outline a multi-faceted approach combining data collection, modeling, and evaluation. Emphasize practical trade-offs and how you would validate the solution in a production system like Snapchat's.

Pro tip: Mention that position bias is often entangled with other biases (e.g., selection bias) and that a combination of techniques (e.g., randomization, inverse propensity scoring, and unbiased learning-to-rank) is usually more effective than a single method.

1. Define and Diagnose

Clearly define position bias and explain how it manifests in ads ranking data (e.g., higher positions get more clicks regardless of relevance). Discuss methods to detect it, such as comparing click-through rates across positions for identical ads.

2. Data Collection Strategies

Describe how to collect unbiased data, such as running randomized experiments where ad positions are randomly assigned, or using propensity scores from logged data to correct for bias.

3. Modeling Techniques

Explain modeling approaches to mitigate position bias, such as inverse propensity scoring (IPS), unbiased learning-to-rank (e.g., using position as a feature during training but not at inference), or multi-task learning that separates relevance and position effects.

4. Evaluation and Validation

Outline how to evaluate the effectiveness of bias mitigation, using offline metrics (e.g., counterfactual evaluation) and online A/B tests to measure improvements in ranking quality and business metrics.

5. Trade-offs and Productionization

Discuss trade-offs between complexity, latency, and performance, and how to integrate the solution into a production system, considering factors like scalability and maintainability.

Key Points to Mention

  • Inverse Propensity Scoring (IPS) and its variants
  • Randomized experiments for unbiased data collection
  • Unbiased Learning-to-Rank (ULTR) and position-aware models
  • Counterfactual evaluation and offline metrics
  • Trade-offs between bias correction and model complexity
  • Online A/B testing for validation

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

Q5

How would you evaluate an ads ranking model, both offline and online?

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

Offline I went with AUC and log loss.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the evaluation goal: offline metrics predict online performance, while online metrics measure actual business impact. Then, structure your answer by covering offline evaluation (ranking metrics, calibration, counterfactual methods) and online evaluation (A/B tests, interleaving, guardrail metrics), emphasizing the importance of aligning offline and online results.

Pro tip: Highlight the limitations of offline metrics and the need for online validation, and mention how you would use counterfactual or unbiased evaluation techniques to bridge the gap. Also, stress the importance of monitoring long-term effects and potential feedback loops in ads ranking.

1. Define Evaluation Goals

Clarify what you aim to evaluate: model accuracy, user engagement, ad revenue, or long-term user satisfaction. Align offline and online metrics with these goals.

2. Offline Evaluation

Use historical data to compute ranking metrics (e.g., AUC, NDCG, MAP), calibration metrics (e.g., predicted vs. observed CTR), and business metrics (e.g., expected revenue). Consider counterfactual or off-policy evaluation to address bias.

3. Online Evaluation

Run A/B tests to measure causal impact on key metrics (CTR, conversion rate, revenue, user retention). Use interleaving for faster iteration and guardrail metrics to detect regressions.

4. Compare and Iterate

Analyze discrepancies between offline and online results to improve offline proxies. Iterate on model features and evaluation methodology.

5. Monitor Long-Term Effects

Track long-term metrics and potential feedback loops (e.g., ad fatigue, user churn) using holdout groups or switchback tests to ensure sustained performance.

Key Points to Mention

  • Offline metrics: AUC, NDCG, MAP, calibration (predicted CTR vs. actual), and business metrics like expected revenue.
  • Counterfactual/off-policy evaluation to mitigate bias in offline data.
  • Online metrics: CTR, conversion rate, revenue, user engagement, retention, and guardrail metrics (e.g., app crashes, latency).
  • A/B testing best practices: randomization unit, sample size, statistical significance, and avoiding peeking.
  • Interleaving for faster and more sensitive comparison of ranking models.
  • Long-term effects and feedback loops: ad fatigue, user satisfaction, and ecosystem health.

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

Q6

What serving constraints matter when deploying an ads ranking model, and how do they affect your architecture choices?

System DesignTechnical Trade-offs
Author's notes

Latency is the main one.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the unique serving constraints of Snapchat's ads ranking system, such as strict latency budgets, massive scale, and real-time bidding. Then, connect each constraint to specific architectural decisions, like model complexity, feature serving, and hardware choices, emphasizing trade-offs between accuracy and efficiency.

Pro tip: Quantify the impact: mention that a 10ms latency increase can reduce ad revenue by X%, showing you understand business metrics. Also, highlight how you'd monitor and adapt to changing constraints in production.

1. Identify Key Constraints

Enumerate the primary serving constraints: latency (e.g., <100ms), throughput (millions of QPS), cost, and reliability. Consider Snapchat-specific factors like mobile device heterogeneity and network variability.

2. Map Constraints to Architecture

For each constraint, propose architectural responses: model distillation for latency, sharding for throughput, spot instances for cost, and fallback models for reliability. Explain how these choices interact.

3. Discuss Trade-offs

Analyze trade-offs between model accuracy and serving efficiency. For example, using a simpler model may reduce latency but lower CTR; discuss how to balance via techniques like cascading or early exit.

4. Propose a Concrete Design

Outline a high-level architecture: two-stage ranking (lightweight retrieval + heavy ranking), feature store with low-latency access, and hardware acceleration (GPUs/TPUs) where beneficial.

5. Address Monitoring and Iteration

Explain how to monitor latency, throughput, and model performance in production, and how to iterate on architecture as constraints evolve (e.g., A/B testing, canary deployments).

Key Points to Mention

  • Latency budget: end-to-end <100ms, with model inference <10ms
  • Throughput and scalability: handling millions of requests per second with horizontal scaling
  • Cost efficiency: using model quantization, pruning, and spot instances
  • Reliability and fallbacks: graceful degradation with simpler models or cached results
  • Feature serving: low-latency feature store and real-time feature computation
  • Hardware acceleration: leveraging GPUs/TPUs for heavy ranking models

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