Don't let this company waste your time. They sent me a coding exercise before even having an initial phone screen, which seemed really odd and frankly kind of rude. It would be nice to be able to find out more about the role and whether or not it's what I'm looking for before spending any time on a coding exercise.

The coding exercise was fairly simple, though, so I did it and sent my notebook in. I never received any acknowledgement from Klaviyo that they had even received my notebook. I followed up a week later and still heard nothing. A month later, I received a generic, auto-generated rejection email.

This was actually the second negative interview experience I've had with Klaviyo. A few years earlier, I had applied to a different role. I had an initial phone screen with HR, then they sent me a complex coding exercise that I spent an entire weekend working on. They said I had advanced to the next round and scheduled a phone interview with the hiring manager. The hiring manager never joined the call. We rescheduled and the hiring manager no-showed that call as well. We rescheduled a third time and the manager was a no-show again. At that point, I withdrew my application.

I do not recommend applying to this company. They have no respect for the people applying to work there and will waste your time. My guess is that if they have this little respect for their applicants, they probably also don't care about their employees. AVOID AT ALL COSTS.

Exercise The attached CSV file lists the customer, date, and dollar value of orders placed at a store in 2017. The actual gender and predicted gender of each customer is also provided.

Complete each of the following activities in a Jupyter notebook using Python. Put your name and email at the top of the notebook and include your name in the notebook file name. Send back only your notebook file, and please do not zip it. Please do not exclude $0 orders.

A) Assemble a dataframe with one row per customer and the following columns:

• customer_id
• gender
• most_recent_order_date
• order_count (number of orders placed by this customer)

Sort the dataframe by customer_id ascending and display the first 10 rows.

B) Plot the count of orders per week for the store.

C) Compute the mean order value for gender 0 and for gender 1. Do you think the difference is significant? Justify your choice of method.

D) Generate a confusion matrix for the gender predictions of customers in this dataset. You can assume that there is only one gender prediction for each customer. What does the confusion matrix tell you about the quality of the predictions?

E) Describe one of your favorite tools or techniques and give a small example of how it's helped you solve a problem. Limit your answer to one paragraph, and please be specific.

For each question, state any considerations or assumptions you made.