Raveen Wijewickrama

Raveen Wijewickrama

Vehicle & Sensing Systems Development Lead | Researcher

The University of Texas at San Antonio

A researcher in the department of computer science at the University of Texas at San Antonio, specializing in mobile sensing, microbility, and privacy. I also stress bake!

Interests
  • Mobile Sensing
  • Micromobility
  • Wearable Systems
  • Human-Computer Interactions
  • AI-Generated Content
  • Privacy
Education
  • PhD in Computer Science, 2024

    The University of Texas at San Antonio

  • MS in Computer Science, 2017

    Wichita State University, Kansas

  • BS in Computer Science, 2015

    Asia Pacific Institute of Information Technology (APIIT) Sri Lanka

Skills

Python
Java

Android

HTML
JavaScript

Node.js

SQL
Data Science

Pandas, Numpy, Scipy, Scikit-learn, Pytorch, Tensorflow

Operating Systems

Windows, Linux, Android

Networking

TCP/IP, DNS, DHCP, VPN

Web

Apache, Nginx, Caddy

Most Recent Experience

 
 
 
 
 
Vehicle & Sensing Systems Development Lead
September 2024 – Present Full Time
 
 
 
 
 
Vehicle & Sensing Systems Development Lead
August 2020 – August 2024 Part Time
 
 
 
 
 
Research Assistant
August 2018 – August 2024 Part Time
 
 
 
 
 
Graduate Teaching Assistant
January 2023 – August 2023 San Antonio
 
 
 
 
 
Engineering Intern
June 2022 – August 2022 Remote

Recent Projects

.js-id-scooters
AI Art
The emerging field of AI-generated art has witnessed the rise of prompt marketplaces, where creators can purchase, sell, or share prompts for generating unique artworks. These marketplaces often assert ownership over prompts, claiming them as intellectual property. This paper investigates whether concealed prompts sold on prompt marketplaces can be considered as secure intellectual property, given that humans and AI tools may be able to approximately infer the prompts based on publicly advertised sample images accompanying each prompt on sale. Specifically, our survey aims to assess (i) how accurately can humans infer the original prompt solely by examining an AI-generated image, with the goal of generating images similar to the original image, and (ii) the possibility of improving upon individual human and AI prompt inferences by crafting human-AI combined prompts with the help of a large language model. Although previous research has explored the use of AI and machine learning to infer (and also protect against) prompt inference, we are the first to include humans in the loop. Our findings indicate that while humans and human-AI collaborations can infer prompts and generate similar images with high accuracy, they are not as successful as using the original prompt.
AI Generated Content
The rise of code-generating Large Language Models (LLMs) has introduced a new threat to the software supply chain: package hallucinations. These occur when LLMs generate non-existent or incorrect package names, leading to a novel form of package confusion attack. This paper presents a comprehensive study of package hallucinations across Python and JavaScript, using 576,000 code samples generated by 16 LLMs with varied prompts. Our findings reveal that hallucination rates average 5.2% for commercial models and 21.7% for open-source models, resulting in over 205,000 unique hallucinated packages. We analyze the factors contributing to these hallucinations and propose mitigation strategies that significantly reduce their occurrence without sacrificing code quality. Our results highlight package hallucinations as a persistent and systemic challenge in LLM-based code generation, warranting urgent attention from the research and developer communities.
ScooterLab
ScooterLab is a National Science Foundation (NSF) funded community research infrastructure initiative at the University of Texas at San Antonio (UTSA). This publicly-available micromobility testbed and crowd-sensing/crowd-sourcing infrastructure will provide researchers access to a community of riders and a fully operational fleet of highly customizable dockless e-scooters. The research team is prototyping infrastructure to be easily retrofitted with state-of-the-art sensors in partnership with collaborators, enabling real-time collection of fine-grained research data from micromobility rides.

Recent Publications

Quickly discover relevant content by filtering publications.
(2025). ScooterLab: A Programmable and Participatory Sensing Research Testbed using Micromobility Vehicles. In PERCOM.

PDF Cite

(2025). We have a package for you! A comprehensive analysis of package hallucinations by code generating llms. In USENIX.

PDF Cite

(2024). MirageFlow: A New Bandwidth Inflation Attack on Tor. In NDSS.

PDF Cite

(2023). SkinSense: Efficient Vibration-based Communications Over Human Body Using Motion Sensors. Internet of Things, 23(1).

PDF Cite

(2022). An Investigative Study on the Privacy Implications of Mobile E-scooter Rental Apps. In WiSec.

PDF Cite