Stripe数据相关面经_Stripe数据相关面试真题及高频题

1.how would you improve it

2.What do you do

3.Shipping Path and Cost Retrieval

4.Catch Me If You Can - Part 3

5.Catch Me If You Can - Part 2

6.Catch Me If You Can - Part 1

7.Multiple Currency Conversions Using BFS

8.Optimal Currency Conversion Rate Calculation

9.Single Intermediate Currency Conversion

10.Direct Currency Conversion Rate Function

11.Designing Metrics and Writing SQL Queries

12.Modeling and Evaluation Case Study

13.Predicting a Product's Target Audience

14.Merge Transaction Records

15.JSON Integration Task

16.Debugging a Python Library

17.Pair Rank

18.Currency Conversion

19.Categorical Feature Encoding

20.Data Profiling and Imputation

21.User Feature

22.Credit Card Redaction

23.Pair Rank/Mutual Rank/Wishlist

24.Server Penalty

25.HTTP Parser

26.Optimizing Money Transfers to Meet Minimum Balance Requirement

27.CSV Data Transformation

28.CSV String Parsing and Object Transformation

29.Stripe Interview Coding Challenge

1. how would you improve it

what are you 3 favorite products, choose 1 and how would you improve it

2. What do you do

You are the GM of Disneyland. You want to lower the length of the waiting lines. What do you do?

3. Shipping Path and Cost Retrieval

Given an input list of strings in the format ['US:UK:UPS:4', 'UK:CA:UPS:3', 'US:UK:FEDEX:4', ...], where each string represents a shipping route with source, destination, carrier, and cost, answer the following questions: 1. For given source (src) and destination (dest), return the shipping path, method, and cost. 2. For a given stop, return the path, method, and cost. 3. For a given stop, return the path, method, and cost with the minimum cost. 4. For unlimited stops, find the path, method, and cost with the minimum cost.

4. Catch Me If You Can - Part 3

Enhance the fraud detection algorithm to allow merchants to dispute transactions marked as fraudulent. If a dispute is accepted, the transaction will not count as fraudulent, and merchants can return to non-fraudulent status if their fraud ratio drops below the threshold. The input includes the same elements as in Part 1 and 2, with the addition of disputes provided in a specific format. Update the list of fraudulent merchants based on accepted disputes.

5. Catch Me If You Can - Part 2

Modify the fraud detection algorithm from Part 1 to use the percentage of fraudulent transactions as the threshold for marking a merchant as fraudulent. The threshold is now a fraction representing the percentage of fraudulent transactions allowed. If a merchant's fraud percentage ever meets or exceeds this threshold, they will be marked as fraudulent and remain so even if their percentage drops later. The input is the same as in Part 1, but with the threshold representing a fraction. The output should list the fraudulent merchants.