Interview Questions

Bini T. - "I understand this is more focused on ML. However, I have a system question. If users allow us to access their location, or they send location via text box, could we use CDNs for the search without hitting our database? We only query the database when we have zero information on location. Other questions: does embedding always guarantee information on location? Do we discharge the user images after we return a prediction? I heard the feedback that we should keep it for future learning. What would"See full answer

Dinar M. - "Even more faster and vectorized version, using np.linalg.norm - to avoid loop and np.argpartition to select lowest k. We dont need to sort whole array - we need to be sure that first k elements are lower than the rest. import numpy as np def knn(Xtrain, ytrain, X_new, k): distances = np.linalg.norm(Xtrain - Xnew, axis=1) k_indices = np.argpartition(distances, k)[:k] # O(N) selection instead of O(N log N) sort return int(np.sum(ytrain[kindices]) > k / 2.0) `"See full answer

Connor W. - "In the expected value of a coupon problem, you calculated variance of a binomial distribution, and used the satandard deviation, square root of variance, to calculate the confidence interval. Will that approach work the same here? For fair coin: (Heads = 0, tails = 1) Var = 10 * (.5)(1-.5) = 2.5 Stdev = Sqrt(2.5) = 1.581 Mean = 5 Z-score = (Observed Val - Mean) / Stdev = (10 - 5) / 1.581 = 3.164 P val = 0.0008% (Slightly different from the video's solution of 0.00097) Pros of this approach: It"See full answer

K v K. - "Not my response, just the Points noted, : Show passion for the work - certifications, competitions, etc., Show the contribution to the Org Has relevant work experience Has the resume been crafted for the job opening and aligns with the job opening Resumes typically get 15-20 seconds for review "See full answer

Abhinav J. - "Use Normalization when: When using pixel values (0-255) into a Neural Network. ➔ Normalize the data between [0,1] to avoid huge input values that could slow down training. When using k-Nearest Neighbors (kNN) or K-Means Clustering. ➔ Because distance metrics like Euclidean distance are highly sensitive to magnitude differences. You are building a Recommender System using Cosine Similarity.➔ Cosine similarity needs data to be unit norm. Use **Sta"See full answer

Karthik M. - "I would always pick a specific product and make a case. Answering this question without a specific product in mind makes the answer unstructured in my opinion. For instance, I would pick Gmail as an example to answer this and structure my response. Clarifications & Assumptions Let us pick a specific product and structure our response - for instance - let us pick Gmail as the product that you want to analyse and make a Go / No Recommendation to Sunset the Product. Overall Google Revenu"See full answer

Akash A. - "not able to understand the accent of the candidate"See full answer

Meenakshi sundaram M. - "Structure: 1. Ask Clarifying Questions 2. Look at external factors 3. Look at Internal factors ( Slice the data aross different cuts and plan accordingly) Clarifying Questions: What's an outbound message? Is it something Linkedin users send among each other? Decline trends: Is the decline steep or gradual? From when, are we seeing this decline (Say since a week, fortnight, month, etc) External Factors: Have we seen any competition led changes/ new campaigns etc? Have we see"See full answer

Rick E. - " from typing import List def getnumberof_islands(binaryMatrix: List[List[int]]) -> int: if not binaryMatrix: return 0 rows = len(binaryMatrix) cols = len(binaryMatrix[0]) islands = 0 for r in range(rows): for c in range(cols): if binaryMatrixr == 1: islands += 1 dfs(binaryMatrix, r, c) return islands def dfs(grid, r, c): if ( r = len(grid) "See full answer

onering2ruleall - "My leadership style is flexible and adaptive, it varies depending on the team members and the needs of the company. My leadership goal is to empower the team and inspire and grow leaders. In order to achieve that, I combine transformational, democratic and coaching leadership styles. Usually when we are facing a new type of challenge, or at the early stage of a project, I like to adapt the transformational leadership which allows me to listen to all the suggestions from the team members and sta"See full answer

Anonymous Roadrunner - "from typing import List def three_sum(nums: List[int]) -> List[List[int]]: nums.sort() triplets = set() for i in range(len(nums) - 2): firstNum = nums[i] l = i + 1 r = len(nums) - 1 while l 0: r -= 1 elif potentialSum < 0: l += 1 "See full answer

Prajwal M. - "from typing import List def maxprofitgreedy(stock_prices: List[int]) -> int: l=0 # buying r=1 # selling max_profit=0 while rstock_prices[l]: profit=stockprices[r]-stockprices[l] maxprofit=max(maxprofit,profit) else: l=r r+=1 return max_profit debug your code below print(maxprofitgreedy([7, 1, 5, 3, 6, 4])) `"See full answer

Dan D. - "I love the answer and framework, my summary of the interview: it is a slightly modified version of Porter's 5 Forces: User benefit -> Bargaining Power of Buyers Apple benefit -> Talks about the company mission and culture (esp. on the relatively closed eco-system) It was expanded more in the latter half of the interview (SWOT + a bit of Value Chain) Competitive landscape -> Industry Rivalry + Threat of New Entrants + Threat of Substitutes Partnership Impact -> Bargaining Power of"See full answer

Anonymous Cat - "Good practice video; however, at minute 6:40 in this video, the interviewer confirmed the app version causes no issue! While the final answer was the app version!"See full answer