School of Technology and Computing
Hosted by CityU’s Center for Cybersecurity Innovation and School of Technology and Computing (STC), the Applied Research Symposium is open to faculty, professionals, and students from any discipline, university, or professional organization. At the end of each quarter, any students who would like to share their projects including capstone courses will present their outcomes. This symposium will provide opportunities for students, researchers, and practitioners to discuss the influence and impact of the applied computing on the future of our planet and our society
Event details
Date: Thursday, March 19th, 2026
Time: 4:30PM-5:30PM PT
Contact: lenghiemminhhieu@cityuniversity.edu
Advisors
DR. Sam Chung
Dean,
School of Technology & Computing
(CS497) (CS680)
School of Technology & Computing
(CS497) (CS680)
Michael Lahman
Professor,
School of Technology & Computing
School of Technology & Computing
DR. Brian Maeng
Professor,
School of Technology & Computing
School of Technology & Computing
MORGAN ZANTUA
Professor,
School of Technology & Computing
School of Technology & Computing
TaeJin Kim
Professor,
School of Technology & Computing
(CS687)
School of Technology & Computing
(CS687)
Sivakumar Visweswaran
Professor,
School of Technology & Computing
(CS687)
School of Technology & Computing
(CS687)
Tomas Huber
Professor,
School of Technology & Computing
(CS497)
School of Technology & Computing
(CS497)
Session Chairs
David Vo
Rajdeep Singh Golan
Harrison Le
Clark Ngo
Thao Bui
Vishal Krishna Kumar
Sumit K C
Namrata Vijay Mali
Presenters
Parmbir Singh
Enhancing Aerospace Supply Chain Resilience: A Predictive Analytics Approach to Mitigate Lead-Time Volatility
Course: CS 497
Description: This capstone project proposes the development of a predictive analytics framework designed to improve supply chain resilience through early risk detection and data-driven decision-making. By leveraging machine learning techniques and structured operational data, the project aims to shift aerospace supply chain management from reactive disruption response to proactive forecasting.
Mikel Ignacio Barajas Martinez
Self-Healing Systems: Automated Incident Response Using Observability Data
Course: CS 497
Description: This research addresses the absence of a comprehensive self-healing framework that can leverage modern observability data to automatically remediate common service failure patterns such as memory leaks, resource exhaustion, cascading failures, and service degradation. The proposed solution is an intelligent self-healing framework that sits between the observability stack and application infrastructure, automatically detecting failure patterns through rule-based algorithms and executing safe remediation actions without human intervention.
Deni Xanat Garcia Hernandez
AI-Assisted Browser Extension for Real-Time Web Accessibility and UX Improvement
Course: CS 497
Description: This project presents an AI-assisted browser extension that detects web accessibility and UX issues in real time and generates context-aware recommendations based on WCAG 2.1 guidelines. Unlike traditional tools that only report violations, the system allows developers to apply suggested fixes directly within live webpages, streamlining accessibility remediation and improving implementation consistency.
Oscar Bernal
Exploring the capabilities and limitations of webcam-based computer vision for interactive desktop systems
Course: CS 497
Description: This capstone project explores the practical capabilities and limitations of webcam-based computer vision for interactive desktop systems using open-source tools. The planned approach was to evaluate a small set of representative tasks (e.g., object recognition, text recognition, and human-centric cues) across selected open-source libraries under consistent conditions.
Ahmad Al-Karzoun
TaskFlow Management System: Improving Sprint Planning Support in Agile Teams
Course: CS 497
Description: This capstone project focuses on developing a full-stack application designed to support Agile project management and improve sprint planning workflows. The system allows authenticated users to manage projects and tasks while providing a structured foundation for future sprint planning, analytics, and workload visualization. The goal is to bridge research on Agile planning challenges with a practical software solution.
Sergio Andres Hernandez Guerrero
A RAG System for Student Notes as a Study Support
Course: CS 497
Description: A RAG (Retrieval Augmented Generation) system focused on helping students store and retrieve their notes easily, to serve as a study support. Using OCR (Optical Character Recognition) models, handwritten notes and digital documents are converted to machine readable text for retrieval and helping students locate a topic later on for consultation, for example for exam preparation or review.
Zsolt Kiss
Predictive Maintenance of Aircraft Landing Gear Using Physics-Informed Digital Twin Data and Machine Learning
Course: AI 687
Description: The study uses physics‑informed digital‑twin landing‑gear data and tree‑based machine‑learning models to diagnose faults and predict remaining useful life, demonstrating effective, event‑level predictive maintenance without time‑series sensors.
Rajesh Kumar Kapparada
AutoChain – Agentic AI–Driven Inventory and Supply Chain Intelligence System
Course: CS 687
Description: This project focuses on building an Agentic AI–driven supply chain intelligence system that predicts product demand and supports inventory and supplier decision-making. Using machine learning models and real operational data, the system forecasts demand, monitors inventory levels, evaluates supplier reliability, and generates dynamic reorder recommendations. The goal is to improve supply chain efficiency by enabling data-driven and proactive inventory planning.
Jeffrey Anderson
Modeling Terry Riley’s In C as a Multi-Agent System
Course: CS 687
Description: This project is an interactive web application that algorithmically performs Terry Riley’s landmark 1964 minimalist score for an ensemble of up to 32 virtual musicians. Each musician independently navigates the piece’s 53 musical phrases using a probabilistic state machine constrained by Riley’s original ensemble coordination rule, ensuring no player falls more than three phrases behind the group. Built with React and the Web Audio API (Tone.js), the system synthesizes multiple instrument timbres in real time and includes a live timeline visualization tracking collective ensemble progress producing a performance that is statistically unique on every run.
Shruti Arvind Kherade
ClaimGuard Copilot: AI-Driven Fraud Detection & Triage Assistant
Course: CS 687
Description: ClaimGuard Copilot is an AI-driven fraud detection and triage assistant designed to improve insurance claim investigations. The system integrates tabular machine learning models with graph-based analytics to detect both individual and coordinated fraud patterns. It incorporates SHAP-based explainability to provide transparent, regulator-friendly insights and includes an LLM-powered copilot that generates natural-language investigation summaries for analysts. The platform also features concept drift monitoring to ensure sustained model performance over time. This end-to-end system enhances detection accuracy, reduces manual review time, and provides investigator-centered decision support in regulated insurance environments.
Reema Ramchandra Kadechkar
Pharmaceutical Inventory Optimization Platform for Healthcare Networks
Course: CS 687
Description: This project develops an AI-driven Pharmaceutical Inventory Optimization Platform designed to reduce medication waste and prevent drug shortages across multiple healthcare facilities. The system combines demand forecasting (Prophet and ARIMA models), a multi-agent framework (CrewAI), and real-time dashboards to predict demand, detect expiration risks, and recommend cost-effective transfers between facilities. By transforming inventory management from reactive to proactive, the platform improves operational efficiency, reduces waste by over 40%, and enhances medication availability to support better patient care.
Megha Narendra Simha
SmartCartAI: A Decision-Centric Agentic System for Inventory Waste Reduction and Pricing Optimization
Course: CS 687
Description: SmartCartAI is a multi-agent AI system that helps grocery retailers reduce stockouts and food waste by turning inventory and sales signals into actionable recommendations. It supports chat-based decision support (reorder, discount, donate/discard), risk and feasibility/cost analysis, and a dashboard for item-level insights—integrating forecasting and structured agent workflows to produce explainable, operational next steps.
Chen Yuan
Enhancing Medical NER Retrieval with LLM and Contrastive Learning
Course: CS 687
Description: The primary objective of this project is to investigate the synergy between Large Language Models (LLMs) and specialized Named Entity Recognition (NER) frameworks within the clinical domain. While medical NER is the fundamental engine for extracting structured intelligence from unstructured clinical narratives, such as electronic health records, research compendiums, and diagnostic reports, it often struggles with the semantic ambiguity and professional terminology inherent in medical discourse.
Puqi Liu
Smart Contract Vulnerability Detection
Course: CS 687
Description: A machine learning pipeline that filters false positives from Slither’s reentrancy vulnerability detection in Ethereum smart contracts. The system combines static analysis metadata with code-level semantic features to classify each warning as a genuine vulnerability or false alarm, helping security auditors reduce manual verification effort and focus on real threats.
Aarthi Ganesan
Design of a Retrieval-Augmented Document Question-Answering System for Organizational Knowledge Management
Course: CS 687
Description: This project develops a secure, retrieval-augmented document question-answering system that allows users to ask natural-language questions over organizational documents and receive concise, citation-based answers, improving information access and reducing manual document review in enterprise settings.
Vidyasagar Neerudi
ClearTrade - Explainable AI Financial Recommendation Agent
Course: CS 687
Description: ClearTrade is an Explainable AI–based financial recommendation system that analyzes stock data using technical indicators, fundamental metrics, and news sentiment. The platform integrates machine learning models to generate Buy, Hold, or Sell recommendations and uses explainable AI techniques to provide transparent reasoning and confidence scores for each decision.
Hemanth Goud Donekena
Healthcare Symptom Insight Assistant
Course: CS 687
Description: This capstone project presents a safety-focused healthcare symptom insight assistant using Generative AI. The system uses Retrieval-Augmented Generation (RAG) to provide educational symptom insights based on trusted medical information sources. It avoids medical diagnosis and includes safety mechanisms such as red-flag symptom detection and explainable responses to improve reliability and user safety.
Ruojie Hao
WebGPU-based Geographic Information System
Course: CS 687
Description: The WebGPU-based Geographic Information System (GIS) project focuses on designing and implementing a high-performance, browser-based spatial visualization platform using the WebGPU API. The system integrates real-time geographic rendering with interactive data analysis within a unified, GPU-centric architecture. By leveraging modern GPU acceleration and structured spatial data pipelines, the project aims to improve scalability, rendering efficiency, and usability compared to traditional web-based GIS systems.
Gurvareddy Padigapati
AI-Based Smart Queue and Appointment Management System
Course: CS 687
Description: This project focuses on developing an AI-Based Smart Queue and Appointment Management System that predicts customer waiting times using machine learning techniques. The system analyzes factors such as queue length, number of staff, service priority, and time of day to estimate waiting times more accurately. It also helps manage both walk-in queues and scheduled appointments. The goal of this project is to help businesses manage queues more efficiently, reduce customer waiting time, and improve the overall service experience.
Ashwin Shastry Paturi
An Agentic AI Orchestrator for Full-Stack Legacy Modernization and Design System Synchronization
Course: CS 687
Description: This capstone project presents SynchroMesh, an agentic AI–based orchestration framework designed to address the design-to-code synchronization problem in legacy frontend systems. The system integrates design tokens from Figma and source code repositories from GitHub using the Model Context Protocol (MCP) and coordinates specialized agents through the Google Agent Development Kit (ADK). SynchroMesh detects design-token drift, recommends governed refactoring actions, and supports human-in-the-loop approval via an interactive dashboard, enabling controlled, explainable UI modernization workflows.
Jared Graham
Evaluating Process Mining Against Process Improvement Frameworks
Course: DS 687
Description: This study examines the relationship between process mining and traditional process improvement frameworks including Lean, Six Sigma, and Theory of Constraints. By analyzing event log data, the project evaluates how data-driven observations of process execution align with or challenge the assumptions underlying these frameworks.
Jan McConnell
Signal Extraction from IMDb Ratings and Metadata
Course: DS 687
Description: The Internet Movie Database (IMDb) provides structured data on film titles, production attributes, and aggregated audience ratings. This project analyzes IMDb data to examine how production-related factors, including cast, crew, and title attributes, relate to observable audience engagement as reflected in ratings and vote counts. SQL is used to construct and validate a relational analytical dataset through targeted joins, filtering, and feature engineering. Python supports exploratory analysis and visualization, and R is used for statistical validation. The project emphasizes interpretation of ratings-based signals within complex, real-world data and demonstrates an end-to-end analytics workflow applicable to decision support and applied data science.
Akhila Maddela
Email Spam Detection Techniques Using Machine Learning
Course: DS 687
Description: This capstone project aims to detect email spam messages utilizing machine learning approaches. This project aims to construct a system that can automatically identify and categorize emails as spam or legitimate (ham) emails. This project will use a significant number of emails and apply machine learning approaches, such as the Support Vector Machine (SVM) method, and preprocessing approaches such as tokenization and stop word elimination. This will enable the system to identify patterns in spam emails, which will then be used to filter emails and ensure the security of email communications.
Schedule
Main Room & Session 01
Session 02
Session 03
Session 04
Session 05
Session 06
Session 07
Session 08
Main Room & Session 01
| 4:30 PM – 4:35 PM | Welcoming session _____________________________________________ | Main Advisor: Dr. Sam Chung Main Chair: Harrison Le MSTeams link: Session 1 |
| 4:35 PM – 4:55 PM | Presenter 1: Parmbir Singh Topic: Enhancing Aerospace Supply Chain Resilience: A Predictive Analytics Approach to Mitigate Lead-Time Volatility _____________________________________________ | Chair: Harrison Le Advisor: Tomas Huber Presentation: 15 Minutes Q&A: 5 minutes |
| 4:55 PM – 5:15 PM | Presenter 2: Oscar Bernal Topic: Exploring the capabilities and limitations of webcam-based computer vision for interactive desktop systems ____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Ahmad Al-Karzoun Topic: TaskFlow Management System: Improving Sprint Planning Support in Agile Teams _____________________________________________ | |
Session 02
| 4:35 PM – 4:55 PM | Presenter 1: Mikel Ignacio Barajas Martinez Topic: Self-Healing Systems: Automated Incident Response Using Observability Data _____________________________________________ | Chair: Sumit K C Advisor: Tomas Huber Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 2 |
| 4:55 PM – 5:15 PM | Presenter 2: Deni Xanat Garcia Hernandez Topic: AI-Assisted Browser Extension for Real-Time Web Accessibility and UX Improvement _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Sergio Andres Hernandez Guerrero Topic: A RAG System for Student Notes as a Study Support _____________________________________________ |
Session 03
| 4:35 PM – 4:55 PM | Presenter 1: Zsolt Kiss Topic: Predictive Maintenance of Aircraft Landing Gear Using Physics-Informed Digital Twin Data and Machine Learning _____________________________________________ | Chair: Clark Jason Ngo Advisor: TJ Kim Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 3 |
| 4:55 PM – 5:15 PM | Presenter 2: Jeffrey Anderson Topic: Modeling Terry Riley’s In C as a Multi-Agent System _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Shruti Arvind Kherade Topic: ClaimGuard Copilot: AI-Driven Fraud Detection & Triage Assistant _____________________________________________ |
Session 04
| 4:35 PM – 4:55 PM | Presenter 1: Hemanth Goud Donekena Topic: Healthcare Symptom Insight Assistant _____________________________________________ | Chair: Rajdeep Singh Golan Advisor: TJ Kim Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 4 |
| 4:55 PM – 5:15 PM | Presenter 2: Reema Ramchandra Kadechkar Topic: Pharmaceutical Inventory Optimization Platform for Healthcare Networks _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Aarthi Ganesan Topic: Design of a Retrieval-Augmented Document Question-Answering System for Organizational Knowledge Management _____________________________________________ |
Session 05
| 4:35 PM – 4:55 PM | Presenter 1: Megha Narendra Simha Topic: SmartCartAI: A Decision-Centric Agentic System for Inventory Waste Reduction and Pricing Optimization _____________________________________________ | Chair: Vishal Krishna Kumar Advisor: Sivakumar Visweswaran Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 5 |
| 4:55 PM – 5:15 PM | Presenter 2: Puqi Liu Topic: Smart Contract Vulnerability Detection _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Ashwin Shastry Paturi Topic: An Agentic AI Orchestrator for Full-Stack Legacy Modernization and Design System Synchronization _____________________________________________ |
Session 06
| 4:35 PM – 4:55 PM | Presenter 1: Ruojie Hao Topic: WebGPU-based Geographic Information System _____________________________________________ | Chair: Namrata Vijay Mali Advisor: Sivakumar Visweswaran Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 6 |
| 4:55 PM – 5:15 PM | Presenter 2: Rajesh Kumar Kapparada Topic: AutoChain – Agentic AI–Driven Inventory and Supply Chain Intelligence System _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Gurvareddy Padigapati Topic: AI-Based Smart Queue and Appointment Management System _____________________________________________ |
Session 07
| 4:35 PM – 4:55 PM | Presenter 1: Vidyasagar Neerudi Topic: ClearTrade – Explainable AI Financial Recommendation Agent _____________________________________________ | Chair: Thao Bui Advisor: Sivakumar Visweswaran Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 7 |
| 4:55 PM – 5:15 PM | Presenter 2: Chen Yuan Topic: Enhancing Medical NER Retrieval with LLM and Contrastive Learning _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: <N/A> Topic: <N/A> _____________________________________________ |
Session 08
| 4:35 PM – 4:55 PM | Presenter 1: Jan McConnell Topic: Signal Extraction from IMDb Ratings and Metadata _____________________________________________ | Chair: David Vo Advisor: Dr. Brian Maeng Presentation: 15 Minutes Q&A: 5 minutes MSTeams link: Session 8 |
| 4:55 PM – 5:15 PM | Presenter 2: Jared Graham Topic: Evaluating Process Mining Against Process Improvement Frameworks _____________________________________________ | |
| 5:15 PM – 5:35 PM | Presenter 3: Akhila Maddela Topic: Email Spam Detection Techniques Using Machine Learning _____________________________________________ |