The following data science interview question is testing for your ability to interpret a 'real life' dataset and quickly extract an insight. This particular question asks for a solution in Python (using Pandas), but it would also be common to solve this in SQL or R.
Below is a table schema for a P2P messaging application. The table contains send/receive message data for the application's users.
table name: user_messaging
time_stamp (integer) # timestamp, epoch seconds
sender_id (integer) # id of the message sender
receiver_id (integer) # id of the message receiver
Question: What fraction of active users communicate with at least 15 unique people on March 1, 2018? You should be able to write complete code to answer this, given just the schema above. Solution will be provided in Python for premium users.
Ace your next data science interview
Get better at data science interviews by solving a few questions per week