cv

MS

University of Colorado Boulder

Computer Science

• Design and Analysis of Algorithms
• Natural Language Processing (NLP)
• Computer Vision
• Data Center Scale Computing
• Operating Systems

BTech

Indian Institute of Information Technology (IIIT), Dharwad

Computer Science and Engineering

Graduate Teaching Assistant

Department of Computer Science, CU Boulder

Graduate Teaching Assistant for the course - "CSCI 3753 - Operating Systems", a junior level computer science course. Served as an instructor for a lecture-based recitation section for about 40 students. Responsible for conducting recitations, grading assignments and exams, and holding office hours.

Graduate Research Assistant

Institute of Cognitive Sciences, CU Boulder

Examined the reasoning capabilities of large language models (LLMs) for spoken language understanding during interpersonal interactions, specifically in job interview responses.

Application Developer

International Business Machines (IBM)

As a Salesforce Developer, worked on full life cycle of user facing applications by developing clean and efficient code, delivering timely solutions and substantially reducing backlogs.

Intern

International Business Machines (IBM)

Worked as a shadow resource to transform the management system for NextEra Energy's Florida Power and Light (FPL) division.

Intern

Indian Institute of Technology (IIT) BHU

Worked on the project - "Design and Fabrication of a Brachytherapy Applicator" which is used in the treatement of cervical cancer.

CUNLP at SemEval-2024 Task 8: Classify Human and AI Generated Text

Association for Computational Linguistics

Pranjal Aggarwal, Deepanshu Sachdeva

Tldsmi: Genetic Algorithm Based Network for Text Localization in Distorted Social Media Images

The Social Science Research Network

Palaiahnakote Shivakumara, Pavan Kumar C, Pranjal Aggarwal, Shubham Sharma, Pasupuleti Chandana, M Basavanna, Umapada Pal

Hope Speech Detection on Social Media Platforms

Chapman and Hall/CRC

Pranjal Aggarwal, Jagrut Nemade, Shubham Sharma, Pasupuleti Chandana, Shankar Biradar and Sunil Saumya

Random Decision Forest Approach for Mitigating SQL Injection Attacks

2021 IEEE CONECCT

Best Paper Award: Pranjal Aggarwal, Jagrut Nemade, Shubham Sharma, Akash Kumar, Kshitiz Michael and Pavan Kumar C

Text Detection in Distorted Social Media Images Using Genetic Programming

• Developed a model to detect text in natural images distorted by continuous sharing on social media
• Used MSER to fetch connected components and developed a 3x3 filter using genetic programming
• Classified the components as text and non-text and used the text components with iterative dilation to create bounding boxes around the text

Android Malware Classification

• Implemented sampling techniques to address the issue of working with large datasets
• Implemented Principal Component Analysis (PCA) for exploratory data analysis and to reduce the dimension since the dataset had large number of columns (9503)
• Trained Artificial Neural Networks (ANN) to achieve an accuracy of 67%

Hope Speech Detection on Social Media Platforms

• Trained a Natural Language Processing (NLP) based system to classify text comments as Hope, Non-Hope speech
• Implemented techniques such as Bag of Words, TF-IDF, Word2Vec for text vectorization
• Used data balancing techniques (SMOTE, ADASYN, etc)
• Intriduced a third label (Neutral Speech) for better classification
• Analyzes and experimented on both the datsasets (old and new) with conventional ML models, LSTMs, and adopted BERT Architecture
• Published as a book chapter in Cyber Crime in Social Media (CCSM), August 2022.

Random Decision Forest Approach for Mitigating SQL Injection Attacks

• Trained a classifier to identify and mitigate SQL injection attacks
• Collected data (malicious queries) and identified features to streamline the data for better insights
• Scored a list of keywords according to their risk factors getting an accuracy of about 95%

IBM iX Interns

Internaltional Business Machines

Featured in the newsletter as one of the best interns hired

Best Paper Award

IEEE

Received Best Paper Award for "Random Decision Forest Approach for Mitigating SQL Injection Attacks

Patent

Indian Patent Office

An Improved HDR Brachytherapy Applicator Apparatus – Filed jointly with others, Ref.No.-P.178.IN, Application No.- 201911044724

Student Board Member

Rocky Mountain AI Interest Group (RMAIIG)

RMAIIG is a community of enthusiastic individuals captivated by the recent advancements in generative AI. This group aims to explore and discuss the latest developments in AI, focusing on tools like ChatGPT, DALL-E, Midjourney, and many others. We also encourage and provide a platform for developers to showcase the cool stuff that they are building with AI tools. Additionally, we examine the impact of these tools on business, education, the workplace, law, entrepreneurship, and society as a whole.

Audiobook Narrator

Bhumi Foundation, India

Recorded course textbooks as audibooks for specially abled students