my friend s experience who got selected....... Algo round: 1. Queue in O(1) solution.. I came up with a method and it was not the most optimal one… and I asked if I shud optimize more.. he sed he was happy with the solution I gave and proceeded to the next question… 2. For eg.. if there is a string abda. I shud reverse it n/2 times.. ie… say 1st rev: adba 2nd rev: (during the second reversal I shudnot rev from beg.. but rev the string between ( i+1,j-1) where i=0,j=n-1in the first case….)ie a db a => a bd a…. so it is abda finally….after n/2 reversal we get the initial string which is abda.. this is a favourable case and I increment the count by 1… The question is give the string length say n=4 how do u find out the total number of strings which will have the favourable condition after n/2 reversals…. I came up with a solution… just think over it.. not a difficult one… just 5mins of thinking will get u reach the right answer….
3. There are 100 pertrol pumps between two points A and B/.. u ve to select 10(say) petrol pumps such that the largest distance between any two petrol pumps shud be minimum…I gave a top down approach.. guess he expected a bottom up approach…. But at the end he was quite happy with my answer… 4. tel me a scenario where i cud use MERGE SORT AND QUICK SORT AND WHERE I CAN USE ONE NOT THE OTHER.. and the differences... MERGE SORT: when there is 1GB of things to be solved we can divide that 1 GB into A*B=1GB , such that each division will have a size of A(MBs)... in this case I cannot bring the entire 1GB into memory.. So i ll bring A(MB) into memory at a tym sort it and keep it and sort all the B divisions and do a B-WAY Merge... which will be easier... But this kinda thing cant be done in quick sorttt.... Quick SORT: if we want to find the Kth smallest v can use quick sort without sorting the entire array but its enuf.. till the pivot elemnt is placed at the kth position... (random select...)...... this cant be applied for merge sort..where v have to sort the entire thick to identify.. kth smallest... the interviewer was happy wit it.. Tech round This round involved all concepts ryt from OS, DBMS, Networks, Data structure, Algorithms, ur projects, a slight touch on testing… Topics covered in each subject: OS: Threads, process… difference.. if threads have more advantages y go for process?? Semaphores, Synchronization, concurrency problems, Second chance algorithm in virtual memory chapter…. DBMS: Second highest salary… TRY TO USE TOP… I TOLD HIM TWO SOLUTIONS ONE WITH TOP AND OTHER WITHOUT TOP … But he emphasized me to use TOP first.. So use that… He gave me a set of tables and asked me write queries Indexing was also an important concept he asked me.. B+ trees properties and how used in DB. Normalization of my tables in my project…. Data structures: Difference between binary trees, hash map, Tries. A scenario where BST has advantages than hash map without any collision. How do we avoid collision in hash Example of a good hash being used…(I told STL..he sed no that’s a different concept.. I then sed…Java hash map…..that was the ryt answer… time for getting.. most of the time O(1) and depending on the inputs and the things it myt vary accordingly ) Tries advantages and wen can it be used….and application of it.. Networks: Layers of OSi.. explanation of each layer ☺ Google.com steps involved About DNS Testing: What do u know abt testing… I explained back and white box testing… alpha n beta testin.. and he sed..he was not proficient in it..n stopped… Algorithms: (besides the algo round) Difference in greedy n dynamic… which is better…. Kinda company oriented questions: advantages and disadvantages of that site… wat I know Abt Advertising online… all this went for around 1.5 hrs…. ☺ … On Aug 6, 1:17 pm, TheMitraBoy <gourab5139...@gmail.com> wrote: > Hi, > Directi visited our Campus in 3rd week of July. It was recruiting for 2 > profiles - System Developer and Developer Operations (Sys Admin) > There was an Apti test to start with. Depending on you performance in the > Apti test , you would be shortlisted in either the Sys Dev or Sys Admin > lists. Remember that there would be sectional cutoffs.Besides the common > part in the question paper, you would have questions targeted towards a > specific profile. Lots of Linux commands for the Sys Admin questions. > In the coding round, Sys Dev were given of finding out the Least Increasing > Sub-sequence. In Sys Admin, we were told to mimic the functionality of the > "find" command in Linux with the "-size" parameter. We were allowed to use > any language of our choice and we were told to use APIs in the particular > languages which make use of system calls (line opendir(), scandir() etc in > C). If you want to get through the coding of Sys Admin, make sure you have > grasped the last chapter (on UNIX Programming) on Kerningham and Ritchie and > tried out the programs there. > > Though i didn't make it to the interviews, i can tell you there were > multiple rounds. Each one exceeding 30 mins. These people can be very > selective. Questions were asked on various topics such as DS, OOP , OS, > Networking etc. > > I hope you find this info useful -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com. To unsubscribe from this group, send email to algogeeks+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.
