Sunday, August 08, 2004

Interviews et. all.

The recent blog concerning interviews seemed to have generated a lot controversy. So I decided to have a separate blog for "interviews".


Now! not many folks would agree with this definition, but I think its true. The fundametal assumption of an interview is mistrust. The employer is simply saying; "I dont entirely trust your resume, I dont entirely trust the kind of education your college has provided. I have decided to check it out for myself"

This is true and can even be considered an axiom. By default, if a company badly wants a specific skillset, you are assumed not to have it unless you can convince otherwise during your interview.

the interview:

Now interviews are done for many positions. The rice shop next to my house interviews guys before allowing them to lift their rice-bags. Then there is the Microsoft interview. Having knocked quite a few company doors in the past, I know a thing or two about interviews. I know a thing or two about what not to do. So pearls of wisdon flow henceforth!

The first question 98% of the time (unless you're facing a slob like me) invariably is " Tell me something about what you have done so far". This is what we call out-question paper. The question is out. The answer to this question in a good interview should decide the future course of questions. Bland answers that start with "I've always wanted to work in your company xyz" This evokes puke. Its like dumpin durin the interview. Nobody always wanted to do anything. You aren't a Nobody but still you didnt aaaaaalways wanted to do anything. Even if you did don't say it like that. use illustrations and events which actually makes the interviewer infer without you spelling it out that you "always" wanted to do stuff.

Usual answers that show you are just about average, start like "I did my engineering in kichadi-gumbalpatti community college of technolgy textiles and science... bla bla bla". When you are answering like this the interviewer is thinking "yaaawn! let me think of catherine-zeta-jones while this fellow is making noises".

One should really come up with a different answer to this question to get the interviewer interested. Ofcourse, if you have a poor product division, keep the marketing division under leash. So! dont hype yourself to make the interviewer think you can run a 4 second mile. Its better to start innovatively with illustrations; for example say something to the effect (in real interview-english) "when I was doing abc course in nth semester, I wanted to know what problems the person who develop xyz concept faced and so I took up efgh book and looked it up. before I knew it I was neck deep inside the subject and voila! my area of interest was born".

This question is best anwered using the above modular template. If you have done enough research about the open job position, replace xyz, abc, efgh etc with appropriate key words from job description itself. It helps if you believe in the system. Sometimes the interviewer asks ridiculous questions. But doesn't happen like that all the time. The problem arises when you second guess the interviewer, trying to create a mental ball-by-ball commentary of what sort of impression he is making out of you. Keep it simple, keep it stupid. Answer the damn question to the best of your ability and move on...

How it works

Okay in some companies the interview atleast appears planned. The HR guy/ Hiring Manager takes care of the planning. A clear plan on who asks what to the candidate is decided. Companies like Microsoft is like playing jeopardy in my opnion. Its double-jeopardy once u clear the Phone interviews .

Like what suresh mentioned in his comment, people ask different questions for different reasons. While some people always ask a very specific questions which they faced the very morning of the interview, a lot of folks (in good companies) have good technique.

For example, I like Microsofts interviewing technique. I have never been successfull but I still like it. Its because its a well defined process allowing the kind of people they want to get in and the kind of people they dont want to not like the interview at all. They make sure most of the time its a case of "grapes are sour".

Good interviews should have problems to solve. Fundametally, what a person does at work is; solve a problem. Languages like C, C++ are merely tools to solve it. So if a person can use both common sense(which includes basic math, communication skills) and some knowledge he claims to know in his resume, to solve problems posed during the interview he is good enough. A 70/100 should be good enough in an interview. Because in work conditions it will improve to 80/100.


There are many pre conceived notions about interview; " xyz are good questions, abc are stupid questions." All these are subjective. This whole blog is subjective. I am right and so are million others who disagree with me.

There is always a small subset of people who get caught with odd interviewing experinces and dont get through for misc reasons but 80% of the time you flunk in an interview because you aren't right for the job (yet). So the best things would be to constantly refine oneself based on interview experience. It is better to find the problem within us and alter it rather than blame it on a faulty interview process and not evolve at all.

Left to me, I would keep the whole process in just testing what the person learnt from his work experience than actually testing specifics (when they arent required). Usually jobs in a divison keep flowing and changing over time, a person is selected based on the general theme of what the org does rather than to finding out how to set a specific bit in a interrupt vector table that would cause a fan motor to whirr in a toilet thats in a galaxy far far away.