DoorDash数据科学面试真题

职位分类
全部
数据科学
数据分析
计算机科学
人工智能
产品经理
BQ
面试题
全部(15)
A/B testing(0)
Statistics(1)
Product Case(11)
Coding(1)
Modeling(0)
高频题(0)
全部(15)
A/B testing(0)
Statistics(1)
Product Case(11)
Coding(1)
Modeling(0)
高频题(0)
1.DashPass
2.Grocery store add search bar
3.Super late的order
4.Dasher extra pay incentive program
5.Increase radius to 10 miles from 5 miles
6.餐馆引入促销功能
7.Growth
8.Grocery delivery
9.Influence of server down
10.Seasonal effect
11.Compare adopters and non-adopters using causal inference
12.Root cause of not working
13.LeetCode 2353
14.两个product合并
15.新的城市sign on新的餐厅
1. DashPass
1. 如果要launch DashPass这个product,如何evaluate success,然后接下来的followup有如果不能做ab test如何衡量success,除了number of orders还有什么别的metrics
 2. 我们launch了这个product后发现第二个月的subscriber流失率高达50%,如何降低这部分customer的流失
2. Grocery store add search bar
如果在Doordash的grocery store的界面上加一个search bar,如何evaluate,看什么metrics,根据你的matrices给到一些follow up

3. Super late的order
如果实际delivery time和expected delivery time相差20分钟那么算是super late,一个月有多少%这样super late的order
 
 有多少dasher 第一次delivery rating是0
4. Dasher extra pay incentive program
Dasher extra pay incentive program: how to increase dasher response rate? if spend $1mm on extra pay, how to build business case and how to measure the benefit?
5. Increase radius to 10 miles from 5 miles
问increase radius to 10 miles from 5 miles
 
 1. benefit
 
 2. how to find a market to run test
 
 3. how to select metrics
 
 4. how to separate treatment vs. control
6. 餐馆引入促销功能
如何衡量是否应该ship让餐馆引入促销的新功能,如何选衡量标准,如何测试,数据回来了如何分析

7. Growth
 Q: one specific acquisition channel's metric dropped, how would you diagnos what's the cause?
 
 Q: website order v.s. download app, what are some tradeoffs?
 
 Q: how to test the hypothesis? how long to read? how to draw conclusion?

8. Grocery delivery
现在要pilot grocery delivery,选了一家大型连锁超市在两个州的10家店(各5家)onboard到DD测试两个月
 
 Q: how do you define success of this pilot? what are metrics and how to define them?
 
 Q: how to test new feature added for grocery delivery?
 
 Q: how would you analyze difference between restaurant and grocery delivery? if some counterintuitive results are observed, hypothesis and analysis plan?

9. Influence of server down
server down了会有什么影响,哪些metric
10. Seasonal effect
 | 用pre-post comparison,怎么处理seasonal effect;
11. Compare adopters and non-adopters using causal inference
 | casual inference比较adopters vs non-adopters:
12. Root cause of not working
 | if the service does not perform ,how to find the root causes;
13. LeetCode 2353
Design a food rating system that can do the following:

  • Modify the rating of a food item listed in the system.
  • Return the highest-rated food item for a type of cuisine in the system.
Implement the FoodRatings class:

  • FoodRatings(String[] foods, String[] cuisines, int[] ratings) Initializes the system. The food items are described by foods, cuisines and ratings, all of which have a length of n.
    • foods[i] is the name of the ith food,
    • cuisines[i] is the type of cuisine of the ith food, and
    • ratings[i] is the initial rating of the ith food.
  • void changeRating(String food, int newRating) Changes the rating of the food item with the name food.
  • String highestRated(String cuisine) Returns the name of the food item that has the highest rating for the given type of cuisine. If there is a tie, return the item with the lexicographically smaller name.
Note that a string x is lexicographically smaller than string y if x comes before y in dictionary order, that is, either x is a prefix of y, or if i is the first position such that x[i] != y[i], then x[i] comes before y[i] in alphabetic order.

 

Example 1:

Input
["FoodRatings", "highestRated", "highestRated", "changeRating", "highestRated", "changeRating", "highestRated"]
[[["kimchi", "miso", "sushi", "moussaka", "ramen", "bulgogi"], ["korean", "japanese", "japanese", "greek", "japanese", "korean"], [9, 12, 8, 15, 14, 7]], ["korean"], ["japanese"], ["sushi", 16], ["japanese"], ["ramen", 16], ["japanese"]]
Output
[null, "kimchi", "ramen", null, "sushi", null, "ramen"]

Explanation
FoodRatings foodRatings = new FoodRatings(["kimchi", "miso", "sushi", "moussaka", "ramen", "bulgogi"], ["korean", "japanese", "japanese", "greek", "japanese", "korean"], [9, 12, 8, 15, 14, 7]);
foodRatings.highestRated("korean"); // return "kimchi"
                                    // "kimchi" is the highest rated korean food with a rating of 9.
foodRatings.highestRated("japanese"); // return "ramen"
                                      // "ramen" is the highest rated japanese food with a rating of 14.
foodRatings.changeRating("sushi", 16); // "sushi" now has a rating of 16.
foodRatings.highestRated("japanese"); // return "sushi"
                                      // "sushi" is the highest rated japanese food with a rating of 16.
foodRatings.changeRating("ramen", 16); // "ramen" now has a rating of 16.
foodRatings.highestRated("japanese"); // return "ramen"
                                      // Both "sushi" and "ramen" have a rating of 16.
                                      // However, "ramen" is lexicographically smaller than "sushi".

 

Constraints:

  • 1 <= n <= 2 * 104
  • n == foods.length == cuisines.length == ratings.length
  • 1 <= foods[i].length, cuisines[i].length <= 10
  • foods[i], cuisines[i] consist of lowercase English letters.
  • 1 <= ratings[i] <= 108
  • All the strings in foods are distinct.
  • food will be the name of a food item in the system across all calls to changeRating.
  • cuisine will be a type of cuisine of at least one food item in the system across all calls to highestRated.
  • At most 2 * 104 calls in total will be made to changeRating and highestRated.
14. 两个product合并
商家或餐馆只利用送餐员网络送货送餐菜单之类不上app,现在决定要把两个product合并,问到底怎么决定这个决定好不好,问现在决定了合并比较好,问怎么roll-out 新的平台 / portal
15. 新的城市sign on新的餐厅
如果你要在一个新的城市sign on新的餐厅,你能collect 任何data你要collect什么data,标准是什么