HOME

My name is Umair Azfar Khan and I am a graduate from Ghulam Ishaq Khan Institute of Engineering Sciences & Technology which is one of the most prestigious institutes of Pakistan. I did my majors in Computer Systems Engineering and graduated on May 29, 2002. Kindly see my curriculum vitae for more details.

Work Experience:

After my graduation I joined Ghulam Ishaq Khan Institute of Engineering Sciences & Technology as a Teaching Assistant as I wanted to give back what I had learned in the past 4 years of my life. I was given the task of improving the standard of programming of the new batch by conducting the two labs CSE - 101 L : Introduction to Computing & Programming Lab and CSE -102 L : Intensive Programming Lab. I am happy to say that I not only conducted the labs and finished the course before time but I also managed to produce a programming manual to help the future students and Teaching Assistants in preparing and conducting the course. A revised edition will be produced after some time so keep visiting this site. Although this manual was the result of accumulating past labs and creating new ones and took me more than 2 months of hard struggle but I think that such information for easily understanding a programming language should not be kept under wraps. Hence the full manual has been provided here.

Apart from these labs, I also had the responsibility of helping teachers during their lectures, conducting tutorials for the students, helping with the convocation management activities and the lengthy process of arranging internships for the junior batch which required continuous cooperation with companies situated all over Pakistan. Such experiences not only helped me in properly managing my daily routine but also helped me learn a great deal from my seniors and colleagues.

Research & Final Year Project:

The Leprechaun Parallel Processing System aims to convert a network of computers into a parallel processing system which takes as input regular sequential code. For this purpose, we analyze dependencies in the code, partition code based on these dependencies, use evolution techniques to optimize these partitions, and finally execute it using El Rewini’s mapping heuristic. The output of our evolution module is a 30 percent improvement over the trivially partitioned code. The Leprechaun successfully converts sequential code into parallel. The research paper on this project that was selected in Techcom - IEE GIK Institute Student Chapter (April 14 - 15, 2002) is present here.

ABOUT

PICS

WORK

LINKS