1.Measuring Success of Lyft's Shared Product.
2.Understanding Statistical Significance and Power in Research
3. Key Metrics for Successful Shared Rides2) Understanding the Cause of a 3-Minute Avg ETA Increase
4.1. Driver Acquisition: ETA Threshold Decision Making2. Rider Attraction: ETA Threshold Determination 3. Analyzing ETA Changes: Identifying Reasons for 3-Minute Increase
5.Optimizing Fuel Costs for Traveling from Point 1 to N
6.Analyzing a 7% Decrease in Weekly Active Users
7.Coupon Usage Probability and Expected Spending: Assumptions and Conditional Probability
8.Metrics for Supply and Demand Balancing in Transportation
9.ETA Time vs. Time Range: Increasing User Ride Probability Explaining P-Value Simply Calculating P-Value for a Binomial Test Deciding Experiment Duration
10.Demand and Supply Metrics: Balancing and Optimizing ETA
11.Analyzing Active User Drop on Dashboard: Troubleshooting Tips
12.Validating Thresholds for Metrics in A/B Testing
13.Metrics for Measuring Supply and Demand
14.Measuring Success for Car-Pooling Service Launch
15.Machine Learning Project Prompt and Troubleshooting Tips
16.Coupon Probability and Expected Value Analysis
17.Diagnosing the 7% Drop in Active Ridership
18.Measuring Demand and Supply in Business
19.Designing Your Experiment for Successful Feature Launch
20.Investigating a 3% Drop in DAU: One-Minute Summary
21.Possible Reasons and Remedies for Driver not Moving after Accepting a Ride Request.
22.Coupon Strategy Design with KPI and User Response Estimation
23.Measuring Success of Lyft's Shared Product.
24.Understanding and Applying Recall and Precision Metrics
25.Understanding the Concept of P-Value.
26.Assumptions of Linear Regression: Explained.
27.Understanding t-test and f-test in Statistics.
28.Random Forest vs Gradient Boosting: A Trade-Off?
29.Introduction to Gradient Boosting Algorithm.
30.Bagging Vs Boosting: Understanding the Differences
31.Measuring User Experience: An Overview.
32.Decode Ways
33.Understanding Statistical Significance and Power in Research
34.Investigating a 3-Minute Increase in AVG ETA.
35.Optimizing the Rider-Driver Match Window Time.
36.Metric-based matching and optimization for riders and drivers.
37.Measuring Success of a Shared Rider (Car Pool) Launch
38.Q1. Layman's Explanation of P-Value Q2. High-Level A/B Testing Workflow Overview
39.Investigating Decrease in New Driver Activation Volume.
40.Implementation of Word2Vec
41.Feature Selection Techniques for High Dimensional Data
42.What is Lasso and How to Use It?
43.Understanding P-Values of Coefficients in Logistic Regression
44.CNN Knowledge and Image Processing Techniques Interview
45.Predicting House Prices: A Case Study
46.Consequences of p>n in Linear Regression
47.P Value in Statistics
48.Understanding Confidence Intervals.
49.Dealing with Price Increases: Investigating Week-on-Week Changes
50.Calculating Binomial Distribution.
51.ETA Adaptation for Supply Surplus and Shortage
52.Diagnosing a 7% Drop in Active Ridership
53.Designing an Experiment to Convert ETA to a Range
54.Choosing Features, Selecting Models, and Evaluating Metrics
55.Optimizing Matching Time in Shared Ride Services
56.What is the Assumption of p-value?
57.Analyzing Reasons for 7% Drop in Customer Numbers.
58.Identifying Reason for 7% Decline in Active Ridership
59.KPIs for Evaluating Launch Readiness of Share Ride Feature
60.Distribute drivers
61.Measure undersupply
62.Select metrics
63.Conditional probability
64.Independence
65.Probability
66.Probability
67.P-value
68.Sample size
69.Eta question
70.AB Test
71.Increase eta reasons
72.Shared ride metrics
73.Strategy case
74.Validation assumption
75.Drop driver reasons
76.Evaluate supply/demand balance
77.Experiement design
78.Drop n % and analyze why
79.OLS的assumption
80.Driver Earnings Metrics Analysis
81.Business Decision Impact Analysis
82.Describe a project you recently worked on.
1. Measuring Success of Lyft's Shared Product.
| Lyft was about to launch a "shared" product, where riders are offered certain discount so that multiple riders going to the same destination can be matched to one driver. This saves riders' cost and increases drivers' potential earnings as well (b/c drivers are paid per hour per mile). But if there is only one rider matched to a driver, then Lyft loses money b/c a discount is offered to the rider. What KPIs would you use the monitor the shared business?
2. Understanding Statistical Significance and Power in Research
| Q1: Explain P-value in a non-technical way
Q2: What is power
Q3: how to comprehend p-value in binomial distribution
3. Key Metrics for Successful Shared Rides2) Understanding the Cause of a 3-Minute Avg ETA Increase
| 1) shared ride 成功的key metric有哪些？
2) 经典考题：问Avg ETA 增加3min的原因？
3) Lyft考虑把ETA改为range (10-15 min), instead of fixed value 问如何设计实验 (A/B test)？4) how to optimize the 'window time' before the rider receives the ETA. 'window time' is defined as the time to match rider and driver.
4. 1. Driver Acquisition: ETA Threshold Decision Making2. Rider Attraction: ETA Threshold Determination 3. Analyzing ETA Changes: Identifying Reasons for 3-Minute Increase
| 1> how to decide the threshold of ETA to acquire more drivers
2> how to decide the threshold of ETA to attract more riders
3> If ETA increase 3 mins this week compared with last week, how to find out the reason
5. Optimizing Fuel Costs for Traveling from Point 1 to N
从点1走到点n，每个点都有一个加油站，对应不一样的price，点与点之前travel需要消耗一定数量的油。最开始在点1的时候没有油。问如果写optimization model minimize 加油的cost。followup：
1. 这个可以什么方法解
2. 如果把问题改成directed graph怎么做。
解锁更多真题，请点此 登录 篱笆帮
高频题（0）
