Monday, September 08, 2008

Interview @ CCIE Lab

"On August 27, Cisco will introduce a pilot for the CCIE Routing and Switching lab exam in Beijing, China. The pilot will add a 10-minute interview that will assess the candidate's ability to apply expert-level networking skills and knowledge to networking problems that are encountered on the job. After the lab orientation, a panel of three experts will conduct a verbal interview with each candidate, asking a series of expert-level networking questions. The ability to correctly answer these questions will affect the exam score. After completing the interview, the candidate will have the complete 8 hours to complete the lab portion of the exam. These scores will then be added up and then combined for a total score which will decide a pass or a fail."

There was a discussion in Internetwork Expert's CCIE Blog with regard to the above announcement. It's not published officially in Cisco website, even I saw it in our internal web as well, and that message is only being sent through email to any CCIE candidate registers for the lab in Beijing. So it's true that they add interview section in CCIE lab. Why only in Beijing? At least for now. And I know most of you have known the answer.

What I want to add here is that the interview concept is actually not something new in CCIE lab. I took the lab before October 2001 and at that time the format was still 2-days exam. So first day is for building the infrastructure, with 45 total points. We need at least to get 30 points minimum to pass to the second day. Then second day morning is for advance features and hell lots of tricks that we may have never seen or heard about it before, and it can give us maximum 30 points. Only if we can get at least 55 points we can proceed to the last section which is troubleshooting.

Within my two attempts with both 2-days format, in two different location Brussels and Tokyo, after the first day the proctor asked each candidate one at a time to get into the lab and sit next to him. Then he would check my work and ask question such as why I did it that way, can I explain in more detail about this technology, is there any other way to do this etc. In Brussels I still remember I even had to do whiteboard presentation to explain the Frame-relay flow control mechanism. But there was no interview for second day morning. Normally after finish up the morning section we had to wait for a bit while since once the proctor found out that we are eligible to continue to the next section, he would start putting the troubleshooting questions into our rack immediately. You know, something likes screw up the cabling, modify IP address, change the password up until the most difficult tricks and usually there were 25 problems that can give us 25 points maximum.

But wait, it's not over yet even after we finish the troubleshooting successfully. My proctor in Tokyo still asked questions about how did I find the problem, which debug commands I run, how did I fix the issue and so on. In Brussels, I needed to find the whole 25 problems to pass and I didn't so the proctor said he doesn't have to check my work. When I took the lab, the interview in both sections, -after first day- and -after troubleshooting- , took much longer than 10 minutes. And my proctor in Brussels that time mentioned even I can get the configuration right but if I can't explain it he would not hesitate to deduct my points.

So as you see, the idea of this interview in the lab is not new. And it should not be something to be afraid of as any CCIE candidate who's ready to take the lab, and willing to pay USD 1400 plus all the travels, should be able to pass such interview easily. I still believe the interview should be done like those old days, so after the candidate finishes the whole lab, to explain why he did the configuration the way he did or just ask for more clarification about particular technology related to the questions. So the proctor can just run the script first to check who's above or at least quite close to the passing point, and only those candidates will have to go through the interview. I don't think those candidates would mind to stay for another hour extra, if that can give them the number they been chasing for. What do you think?


Anonymous said...

Your post let me remember the days worked with tokenring, dlsw+, cat5k, cisco ios debug command reference. And the 2523+2509 2nd-hand routers which paid by my first month salary.

Nickelby said...

I really couldn't agree more with this statement "I don't think those candidates would mind to stay for another hour extra, if that can give them the number they been chasing for. What do you think?".

From what I know from other candidaters and title holders plus going it through myself (although at a very preliminary period at the moment), the dedication, time and determination is a very serious exercise. Hence, an hour just for an interview is not really that much of an issue.

However as pointed out in some other CCIE blogs, it will be a problem if the candidate is not able to speak good English. Understanding English, speaking English and writing English are all three different skills. It would be a pity if some candidate who is an excellent network engineer fail only because his English is not up to an understandable level to the proctor. Just my 2 cents ;-).

Nickelby Thane

Anonymous said...

I definitely like your idea of the proctor validating the thought process at the end of the lab. And, in my own experience in 1 day labs, it still is that way as you go up and ask for clarification of different sections.

Though, from what I can tell, screening questions at the beginning of the lab is going to be used to filter out those people with photographic memories who are just taking the lab to copy it down, without any intention of passing.

While end of lab questions are good, I think that the pre-screening is targeted directly at the problem of labs leaking into the wild.