Interview Questions

Jaswanth P. - " Prioritized Features for Improvement 1. Content Discovery and Recommendation Algorithm Rationale: The algorithm significantly influences user engagement by determining which videos appear on the "For You Page" (FYP). Improving its accuracy can enhance user satisfaction and retention. - Approach: Incorporate more nuanced data points, such as user mood indicators or contextual data (e.g., time of day, trending events) to refine recommendations further. Regular updat"See full answer

Anonymous Salamander - "Microservices are small parts of a application we can deploy them seprately and use them as a application feature."See full answer

Namrata L. - "NPS is the Net Promoter Score. It basically measures if the users will promote or recommend our product or not. Do we have any timelines for this? Also Amazon as in the Amazon MarketPlace right? So to improve NPS, we need to improve the customer experience and keep them more engaged. So lets focus on improving engagement. Mission of Amazon - Help users get any and everything by just clicking some buttons at their doorstep. Users Demographics - Teenagers, Young Adults, Adults, Oldies Pro"See full answer

Vishnu G. - "1. Understand the "Why" (Deep Dive) - Before jumping to solutions, as a PM needs to precisely understand why users are unhappy. NPS gives us a score, but not the reasons. (0 -4 weeks) Analyze Feedback: Go beyond the score. What are Detractors (0-6) saying? What do Promoters (9-10) love? Qualitative Research:(VOC- voice of the customer) Conduct user interviews, analyze support tickets, and observe product usage. Pinpoint specific pain points (e.g., slow p"See full answer

Palak S. - "-- Write your query here With base as( select player_name , team_id , max(scores.gamescore) as gamescore from players join scores on players.playerid = scores.playerid group by playername, teamid) Select team_id , player_name , game_score from ( Select teamid , playername ,gamescore, DenseRank() Over (Partition by teamid order by gamescore desc ) as rnk from base) a where rnk <=2 `"See full answer

Lin R. - "Why can not group by only on name? it gave me incorrect results when I try to do that."See full answer

Rahul J. - "Assumptions: User type: school children living in the USA. Form factor: mobile app on iOS and Android Problem statement School children all across the country visit museums 2-4 times in a year. The museums have info on the contents and layout but that is static. Key value proposition of product is to guide users through the museum Internal company assessment Let's assume the interview is at Google. Google's mission is to organise the world's information so this problem aligns with"See full answer

Jaswanth P. - "Proposed Solution: Awareness and Control Features Algorithm Awareness Campaign Educational Content: Create short videos explaining how the algorithm works to enhance user experience. Transparency Reports: Regular updates on algorithm changes to keep users informed. User-Controlled Engagement Settings Customizable Feed Options: Allow users to adjust content preferences and limit certain types of videos. Time Management Tools: Introduce reminders for screen time limits an"See full answer

Arturo Z. - "function insertItem(heap, item) { heap.push(item); if (heap.length > 0) { let current = heap.length - 1; while(current > 0 && heap[Math.floor(current/2)] > heap[current]) { [heap[Math.floor(current/2)], heap[current]] = [heap[current], heap[Math.floor(current/2)]]; current = Math.floor(current/2); } } } function extractMin(heap) { let smallest = heap[0]; let len = heap.length; if (len > 2) { heap[0] = heap[len - 1]; heap.splice(len - 1); "See full answer

Jyoti V. - "Over-fitting of a model occurs when model fails to generalize to any new data and has high variance withing training data whereas in under fitting model isn't able to uncover the underlying pattern in the training data and high bias. Tree based model like decision tree and random forest are likely to overfit whereas linear models like linear regression and logistic regression tends to under fit. There are many reasons why a Random forest can overfits easily 1. Model has grown to its full depth a"See full answer

Hari priya K. - "Are we talking about Uber app? or generic app? if it is a Uber app, I will proceed further. Uber is a platform where a rider can book a ride from the Uber app. The rider puts their current location or location where they want to take ride from and enters destination address. The Uber app searches for appropriate driver ( based on distance and vehicle) and assigns a driver to the rider. After finishing the ride, the rider and driver were asked to rate each other and on the ride quality . This"See full answer

Googlepm 1. - "Personas : Kids not liking to eat fruits Overweight individuals People who cannot afford fruits Let's select kids who do not want to eat fruits. Pain points : They are finicky eaters They do not know the value of the fruits They do not like the taste of the fruits They want to be fab - not eating healthy alongside friends Solutions : Fruit tracker : self , Global tracker for self + friends, rewards from jamba Juice coupon Recipes : Shown using household it"See full answer

Karthik T. - "It's mainly an experimentation technique for testing new features while the rest of the users are using the old product version of your product. In our case, we were using it for pre-release or announced release features for a specific group of users. We could at any point revert the experience or stop the feature and render the old product version of the product. Based on the success of the feature, we will then do a full rollout of the feature into production. How does it work ? Enable"See full answer