Data Scientist Interview Questions

Aneesha K. - "-- Write your query here select u.userid as userid, IFNULL(sum(purchase_value), 0) AS LTV FROM user_sessions u JOIN attribution a ON u.sessionid = a.sessionid group by user_id order by LTV desc ; Needs a full join. Wondering why cant we do a left outer join here. All the sessions should have complete data."See full answer

Rick E. - " O(n) time, O(1) space from typing import List def maxsubarraysum(nums: List[int]) -> int: if len(nums) == 0: return 0 maxsum = currsum = nums[0] for i in range(1, len(nums)): currsum = max(currsum + nums[i], nums[i]) maxsum = max(currsum, max_sum) return max_sum debug your code below print(maxsubarraysum([-1, 2, -3, 4])) `"See full answer

Alvaro R. - "bool isValidBST(TreeNode* root, long min = LONGMIN, long max = LONGMAX){ if (root == NULL) return true; if (root->val val >= max) return false; return isValidBST(root->left, min, root->val) && isValidBST(root->right, root->val, max); } `"See full answer

Samuel M. - "function main(){ const v1=[2,3, 4, 10] const v2= [3,4 ,5,20, 23] return merge(v1,v2); } function merge(left, right){ const result=[]; while(left.length>0&& right.length>0){ if(left[0]0){ result=result.concat(left) } if(right.length>0){ result=result.concat(right) } return result; }"See full answer

Lucas G. - "I would conduct a sample z-test because we have enough samples and the population variance is known. H1: average monthly spending per user is $50 H0: average monthly spending per user is greater $50 One-sample z-test x_bar = $85 mu = $50 s = $20 n = 100 x_bar - mu / (s / sqrt(n) = 17.5 17.5 is the z-score that we will need to associate with its corresponding p-value. However, the z-score is very high, so the p-value will be very close to zero, which is much less than the standa"See full answer

Aaron W. - "Clarification questions What is the purpose of connecting the DB? Do we expect high-volumes of traffic to hit the DB Do we have scalability or reliability concerns? Format Code -> DB Code -> Cache -> DB API -> Cache -> DB - APIs are built for a purpose and have a specified protocol (GET, POST, DELETE) to speak to the DB. APIs can also use a contract to retrieve information from a DB much faster than code. Load balanced APIs -> Cache -> DB **Aut"See full answer

Moshe S. - "Law is my passion. Traveling all over the world in 5 years"See full answer

Brook - "too much discussing on p-value…. and theoritical things…. country are independant…."See full answer

Alina G. - "SELECT upsellcampaignid, COUNT(DISTINCT trans.userid) AS eligibleusers FROM campaign JOIN "transaction" AS trans ON transactiondate BETWEEN datestart AND date_end JOIN user ON trans.userid = user.userid WHERE iseligibleforupsellcampaign = 1 GROUP BY upsellcampaignid `"See full answer

Sangeeta P. - "Marketing campaigns are run through different channels such as social media, emails, SEO, web advertising, events, etc. Let’s look at some of the overall success metrics at a broader level: Total views for your campaign Unique views for your campaign Returning visitors for your campaign Engagement for your campaign (If it’s a social media campaign, the marketer might be interested in knowing the number of users engaging with the campaign and the type of campaign positive/negative) 5"See full answer

Saurabh N. - "1) Have a common goal 2) Have a clear and fair accountability between teams 3) Ensure conflicts are resolved in time on common issues 4) Promote common Brain-storming , problem solving sessions 5) Most important , Have clear and effective communication established and practised"See full answer

Srijita P. - "Clarification question: How many subscription plans are offered by Tinder ? If there is more than one subscription plan, then we need to ask is the fluctuation happening across all plans or in a particular one ? Assumption: Let's say lower priced subscription plan is showing the most fluctuation and there are only two types of plans In this subscription plan which age group is showing the most fluctuation (18-24,25-30, 30+ etc) ? Is there any seasonality trend observed (eg: placemen"See full answer

Jayveer S. - "WITH suspicious_transactions AS ( SELECT c.first_name, c.last_name, t.receipt_number, COUNT(t.receiptnumber) OVER (PARTITION BY c.customerid) AS noofoffences FROM customers c JOIN transactions t ON c.customerid = t.customerid WHERE t.receipt_number LIKE '%999%' OR t.receipt_number LIKE '%1234%' OR t.receipt_number LIKE '%XYZ%' ) SELECT first_name, last_name, receipt_number, noofoffences FROM suspicious_transactions WHERE noofoffences >= 2;"See full answer

Anonymous Hornet - "To handle the non-uniform sampling, I'd first clean and divide the dataset into chunks of n second interval 'uniform' trajectory data(e.g. 5s or 10s trajectories). This gives us a cleaner trajectory data chunks, T, of format (ship_ID, x, y, z, timestamp) to be formed. For the system itself, I'd use a generative model, e.g. Variational AutoEncoder (VAE), and train the model's 'encoder' to produce a latent-space representation of input features (x,y,z,timestamp) from T, and it's 'decoder' to pred"See full answer