The Interview Loop
Recruiter Screen (30 min)
Standard fit check, behavioral questions, and resume overview.
Technical Loop (3-4 Rounds)
Deep dive into domain knowledge, coding, and system design.
Interview Question Bank
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time you had to pivot a machine learning project because the initial approach failed. How did you communicate this to stakeholders?
#Communication
#Adaptability
#Clear Communication
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time you acted like an owner to resolve a critical production issue in an ML pipeline.
#Ownership
#Incident Management
#Debugging
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time you disagreed with a Product Manager regarding a model's readiness for production.
#Communication
#Stakeholder Management
Machine Learning Engineer
•
Behavioral
•
medium
Describe a time you simplified a complex ML pipeline to improve maintainability and reduce technical debt.
#Refactoring
#Simplicity
#Technical Debt
Machine Learning Engineer
•
Behavioral
•
hard
Tell me about a time you identified a security or privacy risk in an ML dataset. How did you handle it?
#Data Privacy
#PII
#Security
Machine Learning Engineer
•
Behavioral
•
easy
Why Coinbase? And why specifically Machine Learning in the crypto space?
#Motivation
#Crypto Domain Knowledge
Machine Learning Engineer
•
Coding
•
hard
Given a stream of cryptocurrency orders (buy/sell, price, quantity), design a data structure to maintain the order book and efficiently match incoming orders.
#Heaps
#Design
#Trading Systems
Machine Learning Engineer
•
Coding
•
medium
Write a function to detect if a cycle exists in a directed graph representing blockchain transactions between wallets.
#Graphs
#DFS
#Topological Sort
Machine Learning Engineer
•
Coding
•
hard
Given an array of historical prices for a cryptocurrency, write an algorithm to find the maximum profit you can achieve with at most two transactions.
#Dynamic Programming
#Arrays
Machine Learning Engineer
•
Coding
•
medium
Write a SQL query to find all users who made more than 5 transactions within any rolling 10-minute window.
#Window Functions
#Time Series Data
Machine Learning Engineer
•
Coding
•
medium
Given a list of trading intervals [start_time, end_time], merge all overlapping intervals.
#Sorting
#Arrays
Machine Learning Engineer
•
Coding
•
medium
Given a string representing a sequence of crypto ticker symbols, find the length of the longest substring without repeating characters.
#Sliding Window
#Hash Map
#Strings
Machine Learning Engineer
•
Coding
•
medium
Write a SQL query to calculate the 7-day rolling average trading volume for each cryptocurrency asset.
#Window Functions
#Aggregations
Machine Learning Engineer
•
Coding
•
medium
Given an array of integers representing the daily trading volume of an asset, return the top K most frequent trading volumes.
#Heaps
#Hash Map
#Sorting
Machine Learning Engineer
•
Coding
•
hard
Given a stream of prices, design a data structure that supports inserting a price, and calculating the median price in O(log N) or better.
#Heaps
#Data Stream
Machine Learning Engineer
•
Coding
•
medium
Find the Lowest Common Ancestor (LCA) of two nodes in a Binary Tree. (Context: Merkle Trees in blockchain).
#Trees
#Recursion
Machine Learning Engineer
•
Coding
•
easy
Write a SQL query to find the user with the highest total transaction volume in the last 30 days, excluding internal transfers.
#Joins
#Aggregations
#Filtering
Machine Learning Engineer
•
System Design
•
hard
Design a real-time machine learning system to detect fraudulent fiat-to-crypto deposits.
#Fraud Detection
#Streaming Architecture
#Feature Store
Machine Learning Engineer
•
System Design
•
hard
Design a KYC (Know Your Customer) document verification pipeline using Computer Vision and NLP.
#Computer Vision
#OCR
#Pipeline Design
Machine Learning Engineer
•
System Design
•
medium
Design a personalized news and asset recommendation feed for the Coinbase app homepage.
#Recommender Systems
#Two-Tower Models
#Ranking
Machine Learning Engineer
•
System Design
•
hard
Design an ML feature store for Coinbase. How would you ensure consistency between offline training data and online inference data?
#Feature Store
#Data Engineering
#Consistency
Machine Learning Engineer
•
System Design
•
medium
Design a system to predict Ethereum gas fees in real-time to help users optimize their transaction timing.
#Time Series Forecasting
#Real-time ML
#Regression
Machine Learning Engineer
•
System Design
•
medium
Design a system to automatically route incoming customer support tickets to the correct department using LLMs.
#NLP
#LLMs
#Classification
Machine Learning Engineer
•
System Design
•
hard
Design an anomaly detection system for identifying vulnerabilities or exploits in newly deployed smart contracts.
#Anomaly Detection
#Security
#Blockchain
Machine Learning Engineer
•
System Design
•
medium
Design a rate limiter for Coinbase's public API to prevent abuse from high-frequency trading bots.
#Rate Limiting
#Distributed Systems
#Redis
Machine Learning Engineer
•
System Design
•
hard
Design a system to detect account takeovers (ATO) based on user login behavior and device telemetry.
#Security
#Anomaly Detection
#Real-time ML
Machine Learning Engineer
•
Technical
•
medium
How do you handle extreme class imbalance when training an Anti-Money Laundering (AML) classification model?
#Imbalanced Data
#Loss Functions
#Sampling
Machine Learning Engineer
•
Technical
•
hard
Explain how Graph Neural Networks (GNNs) work and how you would apply them to detect malicious actors on the Ethereum blockchain.
#GNN
#Blockchain
#Anomaly Detection
Machine Learning Engineer
•
Technical
•
medium
How would you detect and mitigate concept drift in a model predicting crypto market volatility?
#Concept Drift
#Model Monitoring
#Time Series
Machine Learning Engineer
•
Technical
•
easy
What metrics would you use to evaluate a model predicting whether a user will churn from the Coinbase One subscription?
#Evaluation Metrics
#Churn Prediction
Machine Learning Engineer
•
Technical
•
easy
Explain the difference between XGBoost and Random Forest. Why might you choose one over the other for tabular financial data?
#Tree Models
#Ensemble Methods
Machine Learning Engineer
•
Technical
•
hard
How do you optimize a PyTorch model for low-latency real-time inference in a microservices architecture?
#Model Optimization
#PyTorch
#Inference
Machine Learning Engineer
•
Technical
•
medium
How would you design an embedding generation pipeline for crypto assets based on user trading behavior?
#Embeddings
#Representation Learning
#Word2Vec
Machine Learning Engineer
•
Technical
•
medium
What are the trade-offs between using a generative LLM versus a traditional BERT-based model for classifying transaction descriptions?
#LLMs
#BERT
#Classification
Machine Learning Engineer
•
Technical
•
medium
How do you handle missing or delayed data in a real-time ML pipeline (e.g., delayed blockchain confirmations)?
#Data Engineering
#Streaming
#Robustness
Difficulty Radar
Based on recent AI-sourced data.
Meet Your Interviewers
The "Standard" Interviewer
Senior EngineerFocuses on core competencies, system constraints, and clear communication.
SimulateUnwritten Rules
Think Out Loud
Always explain your thought process before writing code or drawing architecture.