Research Interest: Improving software engineers' tools through data mining, text analysis and machine learning.- Developing automatic techniques to understand context of conversations and extract opinions from software developer chat communications using deep learning architectures.- Developed approaches towards automatically analyzing quality of information in software developer chat conversations using supervised machine learning techniques and natural language analysis.- Conducted an empirical study to learn how software engineers direct their efforts and what kinds of information they focus on within a post selected from the results returned in response to a search query on Q&A forums, especially on Stack Overflow. - Conducted an exploratory study to investigate the potential of developer chat communities as a mining source for software engineering tasks. - Developed a novel approach to identify and locate code examples and the natural language text describing the code segments in research articles. This project also involved designing a tool that takes different kinds of research articles as input and outputs an XML-based representation with markups to associate identified code segments with their specific descriptions.- Conducted an exploratory study to find different types of information embedded in a variety of software-related documents including blog posts, API documentation, code reviews, public chats, etc., and gain insight into the potential value and difficulty of mining the natural language text associated with the code snippets in these documents.- Explored the generality of an approach to automatically determine the high level actions of loop constructs (in Java), based on their structure, data flow and linguistic characteristics.

As sole instructor, I taught the programming course (CISC181) for computer science and computer engineering majors at the University of Delaware during the 2019 10-week summer session. This course covers the principles of computer science illustrated and applied through programming in an object-oriented language (Java), and assumes students have already taken an introductory programming course in either Python or Racket. The programming projects illustrate computational problems, styles and issues that arise in computer systems development and in all application areas of computation.

Assisted teaching in the following courses:Python programming course, SPRING 2015JAVA programming course,FALL 2014Intro to Computer Science with Web Applications, FALL 2015

Domain: Foreign Exchange Trading.- Performed Web-based Testing, Functional and Regression Testing of in-house Foreign Exchange Trading tool, and ensured adherence to quality norms throughout the testing process.- Performed debugging of multiple logs of application workflows for failed scenarios. - Contributed in building the in-house automation test framework using Groovy.

Domain: Computational Advertising. - Single point of contact for the testing team of BRT (Business Response Team) which handled bug/enhancement requests from production. Analyzed software requirement specifications, prepared test plans, test scenarios (using TestLink), created and executed test cases, logged and tracked defects (in JIRA and Bugzilla).- Performed White-box (code-level) testing, Black-Box (Functional) & Regression Testing for the user interface to validate functionalities of different components, created & executed Perl/Shell scripts, ran MySQL queries on the database, and fixed bugs/defects in the source code (in JAVA).- Performed Scoping and estimated the timeline of the project, considering the efforts and number of resources available for a particular production release.

Domain:Online stock/equity trading for the client Charles Schwab. - Developed the automation suite for regression testing (in Perl) and programming /debugging of Perl/Shell scripts to run batch processes in UNIX platform.- Created and executed test cases, identified and logged defects/bugs (Clear Quest).- Performed Sanity testing for new builds & version controlling using Clear-Case.Domain: Multi Functional Peripheral for the client Hewlett-Packard.- Created test plans and test sessions in QC, performed testing on builds - both in development as well as delivery phase using advanced testing tools like HP DSS, HP Web Jet Admin.- Found critical defects through ad-hoc/ exploratory testing, reviewed and fixed bugs in the source code. - Performed build qualification with Smoke & Sanity testing.