About
MY BACKGROUND
I am a Master’s in Computer Science student, enrolled in Intelligent Systems track. I have worked with Financial Firms like Morgan Stanley and JP Morgan in areas of Risk and Collateral Management. I worked as a Software Developer with environments including, JAVA, Python, Scala, Spring Boot, Hibernate, Maven, Jenkins, PostgreSQL, Cloud Foundry, Splunk, Hadoop etc.
My research interest is in exploring Cyber Security with Deep Learning. My projects are related to vulnerability detection in software programs and Smart Contracts, Malware App detection and adversarial attacks on Adaptive Neural Networks.
I am currently working on my Master's Thesis on "Vulnerability Detection using Deep Learning.", advised by Prof. Wei Yang at University of Texas at Dallas.
In my spare time, I volunteer as a Teaching Assistant with Microsoft TEALS for teaching programming to high school students. Whenever time allows, I conduct mock interviews for prospective candidates, preparing for Software Engineering roles under Merit America.
Projects
Common Vulnerability Detection using Deep Learning, Master’s Thesis
advised by Prof.Wei Yang
Training a Deep Learning model on SARD dataset for Java/C# languages, to automatically detect vulnerable code. The focus of the project is to explore similarities in code vectors of the 2 languages and leverage it to build a single model efficient in vulnerability detection across both languages. To our knowledge this is the first attempt which focuses on multiple vulnerabilities (100+) across both languages and explores the common vulnerabilities.
Vulnerability detection in Smart Contracts
along with Maryam Bahjob Imani (PhD candidate at University of Texas at Dallas)
In Review : VLDB 2020
Designed a LSTM and BiLSTM model using TensorFlow, to be trained on bytecode/CFGs of Ethereum Smart Contracts to automatically detect vulnerable Smart Contracts.
ILFO: Adversarial Attack on Adaptive Neural Networks
along with Prof. Wei Yang, Mirazul Haque (PhD candidate at University of Texas at Dallas)
Accepted : CVPR 2020 (pre-print to follow)
With the increase in the number of layers and parameters in neural networks, energy consumption of neural networks has become a great concern to society, especially to users of handheld or embedded devices. In this project, we investigate the robustness of neural networks against energy-oriented attacks. Specifically, we propose ILFO (Intermediate Output Based Loss Function Optimization) attack against a type of energy-saving neural networks, Adaptive Neural Networks (AdNN). An AdNN can dynamically deactivate part of its model based on the need of the inputs to decrease energy consumption. ILFO leverage intermediate output as a proxy to infer the relation between input and its corresponding energy consumption. ILFO has shown an increase up to 100 % in FLOPs (floating-point operations per second) of the remaining FLOPs count of AdNNs with minimum noise added to input images. To our knowledge, this is the first attempt to attack the energy consumption of a DNN.
Education
2018–2020
The University of Texas at Dallas - United States
Masters in Computer Science; Major: Computer Science
Related Courses: Machine Learning, Cyber Security(ML), BigData Analytics, CNNs, NLP, Design and Analysis of Algorithms
2006–2010
Uttarakhand Technical University – India
Bachelor of Technology; Major: Computer Science
Related Courses: Algorithms, OOPs, Operating System, Computer Networks, Database Design.
Experience
Jan 2019 - Jul 2019
PeopleSoft Security Analyst , University of Texas at Dallas
Responsibilities:
-Analyze and perform PeopleSoft account access administration including additions, changes, and deletions using tools like SQL, PSQuery to maintain and monitor security roles.
Sep 2017 - Jun 2018
Associate of Corporate and Investment Banking, JP Morgan and Chase
Responsibilities:
- Developed Micro-services using Spring Boot for supporting complex business modules.
- On boarded Spring 3.x applications for Collateral Triparty Management to an in house Application Monitoring on the cloud (Cloud Foundry).
- Designed a quick build system with Jenkins that saved time during releases.
Nov 2014 - Sep 2017
Senior Associate Platform Level 1 , Sapient Consulting Private Limited
Responsibilities:
- Designed and developed an API to make the project group a golden source of data across the firm. This involved building up systems from scratch that not only could handle any type of data but could make the same available to the end user with bare essential understanding of risk.
- Created REST Web Services and CPS services for enabling better user experience and easy maintenance.
- Developed reconciliation batches from scratch for marking non-modeled trades in time using Java 8, Streams, Multi threading and Spring Integration.
- Maintained a portal for tracing data flow in the entire risk system.
- Built strong integration test suits to prevent applications from potential breaks using Mockito and H2.
Mar 2013 - Nov 2014
System Analyst , Nihilent Technologies
Responsibilities:
- Implemented payment module using Java 8 for healthcare project My Medical Records using Intuit API.
- Migrated production environment from Windows 2000 servers to Linux servers.
- Created Web services with SOAP and REST API for integration with various third parties that improved sales by a good percentage.
- Performed POCs on various tools for improving the application performance, like Code Coverage tools (EclEmma-JaCoCo), Bug Analyzers (SonarCube, FindBug), Build tools (Jenkins) etc.
July 2010 - Mar 2013
Software Developer , CGI Information System and Management Consulting Pvt. LTD.
Responsibilities:
- Designed and developed integration modules for interfaces based on EIP with Spring integration.
- Created processes with Adobe LiveCycle to improve application performance.
- Integrated with LDAP for secondary user authentication
- Designed and developed user authentication module with better usability and reduced manual intervention.
- Innovated features to improve user clarity on the product, like tool-tips with functional description and weight-age of the field.