Data Scientist Interview Questions

Nikita B. - "Mission: Tiktok's mission is to inspire creativity and Joy. Any business wants to make sure that they are serving the value to their customers: For TikTok customers are: Viewers 2. Content Creators 3. Advertisers So few metrics we could measure are: Time spent/day Total no of videos created/day engagement rate = users who interacted in one of the meaningful action on Tiktok / total users at a day level either likes, share, watched vide for at least 5 mins, created video "See full answer

Avon T. - "Initialize left pointer: Set a left pointer left to 0. Iterate through the array: Iterate through the array from left to right. If the current element is not 0, swap it with the element at the left pointer and increment left. Time complexity: O(n). The loop iterates through the entire array once, making it linear time. Space complexity: O(1). The algorithm operates in-place, modifying the input array directly without using additional data structures. "See full answer

Victor N. - "--country names are UPPERCASE but the table in the in the question showing lowercase. That's why it took me a while to figure it out until I ran the country column WITH RECURSIVE Hierarchy AS ( SELECT e.Emp_ID, CONCAT(e.FirstName, ' ', e.MiddleName, ' ', e.LastName) AS FullName, e.Manager_ID, 0 AS Level, CASE WHEN e.Country = 'IRELAND' THEN s.Salary * 1.09 WHEN e.Country = 'INDIA' THEN s.Salary * 0.012 ELSE s.Salary "See full answer

Aarav G. - "Is there a reason a confidence interval was used to solve this problem over just using the mean/expected value directly?"See full answer

Aman M. - "I would want to start by asking a few clarifying questions? What business goal are we trying to achieve with this feature? What decisions will this feature help drive? Is calling feature (audio/video) currently available on 1:1 chats? Does the feature allow for cross-device switch? Once I get some clarity, assuming the goal is to increase user engagement and help drive repeat messenger usage with its user base. I would want to evaluate a few motivation for this feature with looking at"See full answer

Gabriele G. - "DFS with check of an already seen node in the graph would work from collections import deque, defaultdict from typing import List def iscourseloopdfs(idcourse: int, graph: defaultdict[list]) -> bool: stack = deque([(id_course)]) seen_courses = set() while stack: print(stack) curr_course = stack.pop() if currcourse in seencourses: return True seencourses.add(currcourse) for dependency in graph[curr_course]: "See full answer

Laksitha R. - "public static int maxProfitGreedy(int[] stockPrices) { int maxProfit = 0; for(int i = 1; i todayPrice) { maxProfit += tomorrowPrice - todayPrice; } } return maxProfit; } "See full answer

Daniel C. - "select name, stock from products p left join transactions t on p.id = t.product_id order by date desc limit 1"See full answer

Dinesh K. - "i dont know"See full answer

Scott S. - " A couple of years ago, we were working on a project to integrate a new third-party data feed into our existing data processing pipeline. This data feed was critical for enhancing our trading algorithms with more comprehensive market data. Given the tight timeline and high stakes, I decided to push for a rapid implementation. In my eagerness to meet the deadline, I underestimated the complexity of integrating this new data feed. I did not allocate sufficient time for thorough testing and valida"See full answer

Will I. - "Is it bad to get the answer a different way? Will they mark that as not knowing Bayes Theorem or just correct as it is an easier way to get the answer? The way I went is to look at what happens when the factory makes 100 light bulbs. Machine A makes 60 of which 3 are faulty, Machine B makes 40 of which 1.2 are faulty. Therefore the pool of faulty lightbulbs is 3/4.2 = 5/7 from machine A and 1.2/4.2 = 3/7 from Machine B."See full answer

Nidhi S. - "There are couple of reasons for it - Kind of role : Its a product manager role loaded with analytical work, So working with data in stringent regulatory guideline make it more exciting and thrilling. Location & industry is like - Cherry on the cake, Bangalore weather and BFI is at its all time peak as people spending behavior is changing continuously, it will be interesting to see big giants like visa are managing it."See full answer

Yashasvi V. - "WITH RECURSIVE fibonacci_series AS ( SELECT 1 AS n, 0 AS fib1, 1 AS fib2 UNION ALL SELECT n + 1 AS n, fib2 AS fib1, fib1 + fib2 AS fib2 FROM fibonacci_series WHERE n < 20 -- Limit the series to 20 numbers ) SELECT n, fib1 AS fib FROM fibonacci_series ORDER BY n; `"See full answer

Matthew K. - " function climbStairs(n) { // 4 iterations of Dynamic Programming solutions: // Step 1: Recursive: // if (n <= 2) return n // return climbStairs(n-1) + climbStairs(n-2) // Step 2: Top-down Memoization // const memo = {0:0, 1:1, 2:2} // function f(x) { // if (x in memo) return memo[x] // memo[x] = f(x-1) + f(x-2) // return memo[x] // } // return f(n) // Step 3: Bottom-up Tabulation // const tab = [0,1,2] // f"See full answer

Peter W. - "In the question it says: "above the overall average total posts", which to me implying a >, yet in the solution it uses >= Caused me 1 hr to find out. plz fix"See full answer

Christian B. - "with cte as (select ts.employee_id, e.name, t.id as test_id, max(distinct ts.score) as total from test_results as ts join tests as t on ts.test_id = t.id join employees as e on ts.employee_id = e.id group by ts.employee_id, e.name, t.id) select employee_id, name as employee_name, sum(total) as total_score from cte group by employee_id, employee_name order by total_score desc, employee_id asc ;"See full answer