Dataford
Top 33
Prep plan
Updated weekly · Last refresh Aug 30

Amazon DSP Applied Scientist Interview Questions

The questions to prepare for a Amazon DSP Applied Scientist interview. Questions from real interview reports rank first. Updated weekly.

33questions
~5htotal time
Track your progressSign up free to work through all 33 questions and resume where you left off.
Start practicing free →
1
CodingStart here. 8 questions · ~65 min
Binary Search on Sorted ArraysEasy

Explain how binary search works on a sorted array and why its time complexity is O(log n).

time complexityArraysSearchingAmazon DSP
2D Binary Search in a MatrixMedium

Tests algorithm design and correctness reasoning for searching in structured data.

ArraysSearchingMatrixAmazon DSP
More Coding questions with a free account
2
Model Evaluation6 questions · ~49 min
Improve Underperforming Model AccuracyMedium

Approach for diagnosing an underperforming model and improving accuracy through error analysis, feature work, tuning, and bias variance tradeoffs.

Cross-ValidationAccuracyThreshold TuningAmazon DSP
Precision vs Recall TradeoffEasy

Explain the difference between precision and recall, and how each reflects a different type of classification error.

Evaluation TechniquesClassificationConfusion MatrixAmazon DSP
More Model Evaluation questions with a free account

Sign up to see every question

Create a free account to unlock this list and practice real interview questions.

Get my prep plan
3
Machine Learning7 questions · ~57 min
Handling Missing Values in MLEasy

Explain practical strategies for handling missing values in a supervised learning workflow, from diagnosis to modeling and validation.

Cross-ValidationFeature EngineeringRegularizationAmazon DSP
Overfitting and Underfitting in Deep LearningMedium

Explain what causes overfitting and underfitting in deep learning, how to spot each one, and how to reduce them in practice.

underfittingDeep LearningoverfittingAmazon DSP
More Machine Learning questions with a free account
4
System Design7 questions · ~57 min
Design a Real-Time ML Feature StoreHard

Design a shared feature store for training and low-latency inference across many ML systems with strict freshness and consistency needs.

Feature StoreFeature DriftModel ServingAmazon DSP
Design a Travel Recommendation PipelineHard

Design an end-to-end travel recommendation system with retrieval, ranking, feature pipelines, and online feedback loops.

Feature StoreRetrievalRecommendation SystemsAmazon DSP
More System Design questions with a free account
5
Behavioral & Leadership5 questions · ~41 min
More Behavioral & Leadership questions with a free account
The finish line: interview-readyComplete all 33 questions to finish this plan.