SCHOOL OF TECHNOLOGY AND COMPUTING

School of Technology and Computing
Applied Research Symposium
Summer 2025

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

Join Symposium

Event details

Date: Thursday, Sep 11th, 2025
Time: 4:30PM-5:30PM PT
Contact: lenghiemminhhieu@cityuniversity.edu

Advisors

DR. Sam Chung
Professor,
School of Technology & Computing
(CS687)
TJ Kim
Associate Faculty,
School of Technology & Computing
(CS687)
Ahreum Ju
Associate Faculty,
School of Technology & Computing
(CS497)
Michael Lahman
Michael Lahman
Professor,
School of Technology & Computing
(CS497)
brian maeng
DR. Brian Maeng
Professor,
School of Technology & Computing (DS687)
Session Chairs
Ana Karina Balcazar Sanabria
ClarkNgo
Clark Ngo
Namrata Vijay Mali
Honorine Ndom Ndzah
Megha Narendra Simha
Ashwin Shastry Paturi
Kina Anh Bui
Anh Bui
Harrison Le
Rajdeep Singh
Presenters

Jacob Bartley

Cloud-Native JWKS Endpoint for Securing FHIR APIs

Course: CS 497

Description: This project develops a cloud-native JSON Web Key Set (JWKS) endpoint using Azure Functions and Azure Key Vault to secure healthcare APIs built on the FHIR standard. By leveraging OAuth 2.0 and Zero Trust principles, the solution aims to automate key publishing and retrieval, reduce operational overhead, and enable real-time token validation. Current progress includes deploying an initial JWKS function to Azure and refining supporting documentation and presentation materials.

Khalid Hasan

MicroVerse – An Educational Reading Platform

Course: CS 497

Description: Reading scores have been on a downward trend, and the COVID-19 Pandemic has only made the problem worse than it was. Younger children, from first grade to fourth, have been especially hit hard, with children lacking support at home being affected disproportionately. This project explores this problem and offers a solution using Modern Web Development techniques.

Modou Jaguraga

Developing a Framework for Third-Party Risk Assessment and Compliance Monitoring

Course: CS 497

Description: This project focuses on designing a vendor risk assessment and compliance monitoring framework that integrates standards such as NIST SP 800-161r1 and ISO 27001. The framework applies agile Scrum methodology, dynamic vendor risk classification, and compliance checkpoints to ensure continuous oversight, audit readiness, and enhanced supply chain security.

Muhammad Ahmed

SafeScanQR: A Threat Detection Tool for QR Code URLs

Course: CS 497

Description: SafeScanQR is a security tool designed to detect and prevent malicious threats hidden in QR code URLs. It scans, analyzes, and verifies links embedded in QR codes to protect users from phishing attacks, malware downloads, and fraudulent websites. The goal is to provide a safer way for people and organizations to interact with QR codes by ensuring that every scanned link is trustworthy.

Jinquan Lin

A Root Cause Analysis of Negative Product Reviews to Improve Customer Satisfaction

Course: CS 497

Description: This project applies aspect-based sentiment analysis and BiLSTM models to negative Amazon product reviews, performing root cause analysis to identify complaint themes such as quality, shipping, or usability.  The system transforms unstructured feedback into actionable insights to improve customer satisfaction and business decision-making.

Thai Do

Computer Vision in Locating Products on Shelves for Delivery Drivers: Gig Vision

Course: CS 497

Description: How can computer vision help find items in a shop in a shopping order?

William Forde

AI Digest

Course: CS 497

Description: Improving AI summary accuracy using citation requests.

Brooke Erin Tischer

MovieMuse: Enhancing Movie Discovery Through Social Features and Watchlist Tracking

Course: CS 687

Description: MovieMuse is a full-stack web application built with modern web and cloud technologies to provide real-time tracking, social features, and personalized recommendations for movie enthusiasts. 

Sarayu Kancharla

AI Powered College Finder Dashboard for smarter education program discovery

Course: CS 687

Description: The proposed project will create an AI-based decision support tool that will enable students to make a more customized, transparent, and available choice of graduate program. It connects web scraping, semantic search, and LLM-based agents to access and process program data, compare them to student profiles, and produce recommendations with understandable justifications. The interactive Streamlit dashboard not only allows the students to filter and compare the programs where they can best fit but also helps them understand the reasons as to why a certain program fits their preferences, making the otherwise overwhelming task of graduate program selection much clearer and data-driven.

Rashmitha Mothe

Credit Card Fraud Detection with MLOPS integration

Course: CS 687

Description: This project focuses on building a machine learning system to detect fraudulent credit card transactions in a highly imbalanced dataset. The solution combines a backend for data preprocessing and model training with a Streamlit web application that allows users to upload data, train and compare multiple models, generate predictions, and manage saved models. By integrating MLOps practices such as model versioning, evaluation, and promotion, the project demonstrates an end-to-end workflow from data analysis to deployment.

Hai Nguyen

TheChromeCollective - Price Tracker & AI-Powered Marketplace

Course: CS 687

Description: The Chrome Collective is a mobile application that brings transparency and accessibility to the luxury resale market, focusing on Chrome Hearts products. It enables users to track retail prices, browse authenticated listings, and contribute product entries with descriptions and images. By combining price tracking with community-driven features, the platform creates a secure and trustworthy space for collectors and buyers.

Sathiyashivani Sathish Kumar

Smart Recipe AI: Personalized and Secure Recipe Recommendation Platform

Course: CS 687

Description: Smart Recipe AI is an intelligent and secure recipe recommendation system designed to provide personalized meal suggestions based on dietary preferences, health goals, and allergy constraints. By leveraging Artificial Intelligence (AI) and Natural Language Processing (NLP), the system dynamically adapts to user feedback and ensures privacy-aware data handling, supporting healthier and safer eating habits.

Kritthika Shanmugam

AI-Driven Tutor for Driving License Preparation: DriveSmart

Course: CS 687

Description: DriveSmart is a mobile learning platform designed to help beginner drivers prepare for license exams. It integrates a Retrieval-Augmented Generation (RAG) engine with the Ollama Mistral 7B model to deliver accurate, state-specific guidance. The system provides interactive quizzes, chatbot support, and personalized feedback based on performance.

Jiayi Sun

A Smart Document Agent - AskDocAI

Course: CS 687

Description: AskDocAI helps users engage with long documents more efficiently. By combining Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs), the system enables summarization and interactive Q&A directly from PDFs or web links. It provides grounded answers with references to the source material, making the learning process more interactive and focused.

Benjamin Hurst

A Cloud-Native Event-Driven Pipeline for Intraday Analysis of SEC Filings

Course: CS 687

Description: This project builds a low-latency, cloud-native workflow that monitors the U.S. Securities and Exchange Commission’s “Latest Filings” page for fresh insider-purchase disclosures. Each filing is enriched with live price and liquidity data from a market-data API and immediately executes fixed-risk bracket orders through a paper-trading brokerage interface. A built-in back-testing engine replays recent months of history to fine-tune the entry filters and risk-reward levels, with results displayed in a web analytics dashboard, showing how modern cloud tools can transform public regulatory disclosures into actionable, real-time trading insights.

Sriram Bairi

AI-powered Personalized Fitness Tracker with Daily Wellness Generator - FitGenie

Course: CS 687

Description: My project is an AI-powered Personalized Fitness Tracker with Daily Wellness Generator that integrates real-time sensor data, behavior modeling, and generative AI to create adaptive wellness experiences. Unlike traditional fitness apps that only track basic metrics like steps or calories, this system leverages Google ADK sensors for biometric data, Generative AI for customized workout and wellness plans, and the MCP protocol to analyze behavior and adjust recommendations dynamically. The solution eliminates the need for a human coach by providing real-time, personalized feedback loops, improving motivation, consistency, and outcomes. It demonstrates innovation in health tech, AI, and user-centered design.

Wen Yuan

Customer Churn Analysis in E-commerce Using Machine Learning

Course: CS 687

Description: This project explores customer churn prediction in e-commerce using machine learning techniques. By analyzing customer demographics, behavioral data, and engagement metrics, I build and evaluate predictive models to identify churn risks and provide actionable insights for improving customer retention.

Priyanka Jaina

SmartSpend: An AI-Powered Personal Expense Tracker

Course: CS 687

Description: SmartSpend is a web-based application designed to help individuals manage their personal finances efficiently. It enables users to track income and expenses, visualize financial data through interactive dashboards, and plan budgets effectively. Powered by generative AI, SmartSpend also provides real-time personalized financial insights and advice through an integrated chatbot, making money management smarter and more accessible.

Ean Pfeiffer

Evaluating Early-Game Specialization in Multi-Agent LLMs for Strategic Boardgames

Course: DS 687

Description: This project tests whether early-game specialization improves multi-agent LLM performance in Settlers of Catan. Built on a Catan Simulator, I reproduce research that creates a multi-agent LLM pipeline that plays, strategizes, and generates its own code after each game. In this pipeline, I insert an early game specialist that utilizes top level theory, which will handle an early but crucial part of the game.

Sc0tt Edward Taylor

Risk Factors for Idiopathic ILD

Course: DS 687

Description: The primary goal of the current research project is to establish a foundation for identifying promising correlations between a diagnosis of ILD or IPF and clinical factors from a patient’s history.

Sara Verkiyani

Designing a Patient Portal for Scheduling and Supplement Ordering

Course: CS 680

Description: My internship at Comprehensive Medical Center focused on developing a patient-facing web portal that streamlines appointment scheduling and supplement ordering for the naturopathic dispensary. The system design includes role-based access for patients, reception staff, and doctors, along with integration planning for inventory tracking and QuickBooks. This project combines user experience design with backend workflows to improve efficiency and patient engagement.

Vidyasagar Neerudi

Cloud-based E-Commerce Store

Course: CS 680

Description: My internship project involved developing the frontend of a Cloud-Based E-Commerce Store. I focused on creating a responsive and user-friendly interface for product browsing, authentication, and checkout, ensuring seamless interaction and integration with backend and cloud services.

Teja Sree Kokkanti

Elariis inc

Course: CS 680

Description:  During my internship at Quadrant Technologies, I developed a full-stack Company Portal “Elariis” using Python Flask for the backend and MongoDB for database management. The portal included modules for HR leave management, employee directory with CRUD operations, an AI-powered chatbot integrated with the OpenAI API, and analytics dashboards built using Chart.js. The project was deployed using Azure DevOps pipelines for CI/CD. I also implemented JWT-based secure authentication and performed UAT and scalability testing, resulting in a robust, cloud-ready enterprise application.

Veronica Misty Elze

EQīLevel: Emotion-Aware Reinforcement Learning for Multimodal Academic Tutoring

Course: AI 687

Description: EQīLevel is an emotionally adaptive AI tutoring system that combines reinforcement learning (RL), GPT, and real-time emotion detection to personalize instruction. By adjusting lesson difficulty, pacing, and tone to diverse learner’s performance, style, and emotions, EQīLevel enhances engagement and fosters a positive learning experience. This adaptive approach improves both the quality and speed of individualized instruction, enabling students to progress more effectively, confidently, and with greater encouragement.

Schedule

Contact Us

lenghiemminhhieu@cityuniversity.edu