Mbit Wireless Associate Engineer Interview Questions and Answers

Round: Test
Experience: Adaptive DI test. Level of difficulty less than CAT.

Round: Case Study Interview
Experience: This is a very simple case interview. The case was as follows: Client is a micro-lender, they loan funds from a big bank @ 6% SI and loan out the money to farmers at 10% and 20% SI. The segmentation is 60% and 40% respectively. Calculate profit/customer. What is the profit if 2% customers default (uniformly from both segments) on the loan? (Ans: loss incurred)
Tips: This is a very simple round, just stick to the essentials of keeping your structure + calculations neat & clean. Communicate well. You'll get through.

Round: Case Study Interview
Experience: Case: We are a telecom company, planning to launch a new subscription plan. Decide if we should go ahead or not. The plan is as follows: We charge the customers $0.99 for the first 20 minutes and $0.08/minute thereafter. The cost incurred by the company is $0.07/minute. Should we go ahead with the plan?
You will be asked to draw a graph between Profit and minutes spent on a call. To come to a decision you might ask the interviewer for some more data like the distribution or the average time spent by a customer on a call. But none of this is information can be used to decide if we should go ahead with the plan. Unless you have a distribution of percentage of customers vs time spent on an avg call, you can not decide the fate of the plan.
On our campus, almost everybody answered a yes(go ahead) or a no for the plan and still got rejected. I think the actual answer is "None of the above. With the present information nobody can predict if the plan will be profitable in the long run.
Please do not say we can use historical data to come to a conclusion, I did, the question he then posed to me is can we really use historical data to predict the usage pattern of this plan by our customers?? The answer is no, we can not use historical data, as the customers will have an incentive to speak for close to 20 minutes per call.
I also looked at the problem from a customers point of view to check if a customer has an incentive to pick this plan over the others. I asked for the cheapest plan/s out in the market currently, so the interviewer told me that the cheapest plan cost the customer $0.08/minute.
After getting this data, I looked at intersection of the time band at which it becomes cheaper for the customer to pick our plan over the 8 cents plan and the time band which is profitable for us.
I came to the conclusion that the band is very narrow to be profitable and that I do not think that it will generate net profits. So we should not go ahead with the plan.
I assumed that there wont be a large chunk of customers in the above mentioned time band. This is what I think my mistake was.
I believe there is no way of definitely saying whether we should go ahead or not. Hence the answer must be Cant be Determined.
FYI: I also suggested him to launch it in a small region/zone so as to figure out the distribution of customers. No good.

Skills:
College Name: IIT Kanpur
Motivation: Chill out workplace. No pressure of work. But no future either :P wont land you in a great B-school. This is just an office opened by a recently grown American company. They have an office in India just to show the Investors that they plan on expanding and are investing in India and Philippines. I don't think the real work happens here. This might sound like a rant, but you should go ahead and prepare for this company. Its one of the best options straight out of college. Or to settle down or if you like to chill a lot....