I'm a software engineer at Nevelex. I recently graduated from Colorado State University with a masters degree in a field of Computer Science. I currently work on resolving defects from DirecTV set-top box software, in C++ and JAVA.
I have experience in different fields of Computer Science such as Web development, Computer Networking, and Bioinformatics. Also, I was a Teacher once.
My latest past time is Node.js. I created a simple app with Node.js, OpenChat. I'm hoping to improve it over the period of time or create something else altogether.
I love cooking, send me your recipe and I will send mine.
In Spring of 2014, I got an opportunity to join Netsec group at CSU. I worked on real-time BGP routing information monitorig tool called BGPmon. Apart from the regular bug fixes, I delivered a translator module (MRT to XML). I created an tool to simulate and visualize attack events (like NTP reflection attack). I also built a project website.
I started my graduate school in January 2014 to study Master of Science in Computer science at Colorado State University. CS557 (Advanced Networking), CS545 (Machine Learning), CS535 (Big Data), and CS548 (Bioinformatic Algorithms) are some of the subjects that I enjoyed studing most in last one year.
After completing undergraduate study, I got a wonderful opportunity to join Cognizant Technology Solutions. During a period of 30 months, I was engage in number of projects mainly focusing on mobile application, website and emailer development. I learnt numrous things, starting from a client interation to a best codding practices.
I worked with Sagar on this project. We proposed a new method to find best restriction enzyme triad from ~11 million combinations. We implemented the algorithm in hadoop using python to distribute the task of finding shared sub strings in suffix tree. We used MapReduce because of 3V nature of given problem. We also updated the Dell Zhang’s generalize suffix tree implementation to support Unicode characters of large range.
Implemented linear (LLS) and non-linear models(Neural Networks) to predict the demand for bike sharing system
Code implementation was in python and the results were submitted to kaggle for competition
Showed that non-linear model performs much better for this problem
Implemented a sparse de Bruijn graph for genome assembly from scratch in C++
Performed in depth theoretical investigations into various succinct data structures for de Bruijn graph optimization
Project involved development of peer to peer file sharing network that uses protocol resembling BitTorrent.
Implementation of socket programming, fork, event loops was in C language.
Project was implemented using CCNx library from ccnx.org in C.
Compared performance of the protocol in IP vs NDN configuration and found that NDN gave much better performance.
