Logan Coker, Kieran Gilpin, Elijah Palmer, Andrew Rupp, Cooper Wooten
John Lambert, Security Fellow, Corporate Vice President, Deputy CISO @ Microsoft
Memorabiliacs is a multipurpose memorabilia collection management software focusing
on customizability. Most all existing solutions cater toward those who see collectables
as investments and exclusively wish to track the market value over time. Casual collectors
are left with very limited, poorly constructed softwares that require a subscription
to keep track of your physical property. Most existing solutions also lack the ability
to track collections of varied items and focus on one or two types of items.
Memorabiliacs is structured as a web app hosted on a website, users can build and
create their own collections that reflect their real ones.
Our ideal results are having real users populate their collections and accounts as
a live demo and proof of concept for us to display to the senior design committee.
Potential next steps would include investigating required actions to expand the scale
of the applications to a much larger user base.
Sadie Ann, Isabella Breaux, April Gauthreaux and McKinley Humble
Dr. Brad Glisson
As the digital world continues to expand, younger generations increasingly need cybersecurity literacy, yet education in this area struggles to keep pace with their growing exposure to online risks. Cybersecurity concepts are often taught primarily to upper-level students, leaving younger audiences without a strong foundational understanding and more vulnerable to digital threats. Escape the Cyberspace addresses this gap by introducing younger learners to cybersecurity concepts in an engaging and accessible way, while also sparking their interest in the field and demonstrating what cybersecurity involves in practice. This is achieved by transforming traditional lecture-based instruction into an immersive, narrative-driven virtual reality (VR) escape room experience. The final product is a functional VR escape room game that provides a fun, high-engagement environment for learning foundational cybersecurity skills.
Kaylee Matic, Joel Porter, Aidan Schaubhut and Fayrouz Zeidan
Frosty Factory of America, Inc.
Joshua Wheat, Engineer
Manufacturing Execution Systems (MES) are critical to modern industrial operations, bridging enterprise planning software and shop-floor machinery. Frosty Factory of America lacked a centralized system to monitor production workflows, track machine states, and manage operational data in real time. To address this, a full-stack MES was designed and implemented using Go for the backend API, PostgreSQL for data persistence, and Docker for containerized deployment. A flexible Entity-Attribute-Value (EAV) schema supports dynamic machine configurations without schema migrations as equipment or process parameters evolve. The resulting system provides real-time production line visibility, dynamic configuration of heterogeneous machine types, and structured work order lifecycle management. Containerized deployment ensures portability across development and production environments. This project demonstrates that lightweight, open-source tooling can deliver enterprise-grade manufacturing visibility without the cost of commercial platforms. Future work could incorporate predictive maintenance alerting and ERP system integration.
Steven Alleman, Brandon Gros, Austin Hague and Reese Haydel
Dr. Zakaria El-Awadi
Our project involves developing a gamified health and fitness Android application
that allows users to better themselves by completing wellness tasks for screen time
on their phones while also keeping track of progress in fitness and nutrition. This
is important to the field of computer science as it explores finding a solution to
a real-world problem through software development. The problem is that most people
have neglected their physical health and increased screen time due to daily stress.
Our application offers a solution by allowing users to have an engaging way to achieve
screen time by completing daily wellness tasks. It also offers tools to develop and
show progress of nutrition and fitness regimens tailored to the user. Our approach
involves developing a modular UI, goal-setting features, storing user data to tailor
progress and regimens towards them, and gamifying completion of daily tasks. The most
significant outcome so far has been developing a fully functional Android app as all
members of our group are achieving this for the first time. Overall, this project
shows that software development can be utilized as tools for achieving self improvement.
Cade Barfield, Brady Burns, Tyler Johnson, and Bobby Moore
Dr. Kyle Prather
Flexfolio is a multi-portfolio platform designed to improve how professionals present their work across different audiences. In fields such as software development, design, and academia, individuals often need tailored portfolios but lack tools that balance customization with simplicity. Current portfolio builders require duplicated content or complex site management, creating inefficiencies when showcasing the same work in different contexts. This project addresses the need for a system that enables reusable project content across multiple structured portfolios. Flexfolio was developed using a block-based architecture that separates projects from portfolios, allowing users to create once and display content in multiple formats through templates and themes. A functional prototype was implemented and tested to validate content reuse and multi-portfolio configuration. Testing confirmed that projects can be dynamically reused without duplication while maintaining design flexibility. Flexfolio reduces maintenance effort and enables more strategic professional presentation, with future development focused on collaboration and analytics features
Isaiah Hinds, Dylan Johnson, Sarah Wallen, Javen Wilson
Dr. Zakaria El-Awadi
Modern GPS navigation applications often fail to provide real-time, localized
updates for unannounced road hazards such as potholes, debris, and sudden
blockades. This lack of timely data compromises driver safety and reduces
community awareness. The Curby Smart Dashcam addresses this issue by
combining artificial intelligence with a community-driven mobile platform to
deliver accurate, up-to-date road condition reporting. Curby consists of an NVIDIA
Jetson Nano loaded with our own trained computer vision model equipped with a camera,
capable of detecting road hazards in real time. When hazards are identified, geolocation
data is captured and transmitted to a backend server through secure API endpoints.
The device is accompanied by a mobile application for iOS and Android that detects
hazards on an interactive map, allowing users to view and manually report incidents
within their area. Prototype testing demonstrated successful real-time hazard detection
and map synchronization between the device and mobile app. Integrating AI-powered
detection using datasets from Mapillary and RDD2022, Curby enhances driver awareness
and contributes to safer roadways. Future work includes expanding
the hazard classification model and optimizing detection accuracy under varying
weather and lighting conditions.
Donovan Dempster, Drew Ferrington, Caleb Reed and Henri Tran
Altech Solutions LLC
Tom Bowman and Jason Robinson
This project delivers a plug-and-play add-on module for The Little Engine (TLE) that simulates data center sensors and produces realistic synthetic telemetry for safe testing and training. Facility sensor feeds are often restricted due to operational risk and confidentiality, making it hard to validate monitoring workflows and analytics without impacting a live environment. We address this by building a safe practice data center that generates power and cooling sensor data while remaining isolated from real systems. The module outputs continuous fake readings that mimic typical device behavior and can be configured to simulate abnormal conditions and edge cases. When connected to TLE, the add-on watches the simulated stream, highlights energy-use efficiency trends, and provides early warnings when patterns suggest equipment may be nearing failure. It also flags unusual telemetry that could indicate a potential cybersecurity breach affecting data center infrastructure. The result is a self-contained attachment where users view telemetry, trends, and alerts in one place, combining health and security monitoring for a building’s power and cooling systems.
Kelsie Culbert, Jordan Myles, Preston Parker and Savannah Stokes
Catfish Alley Studio
Josh Gillis
Catfish Alley Playspace is a mixed-reality (MR) application for Catfish Alley Studio that aims to make guitar lessons more interactive and approachable for their beginner students. Developed in Unity, this project incorporates hand-tracking and a virtual fretboard overlay, alongside the essential beginner lessons taught by Catfish Alley Studio, to provide the user with visual aids and instant feedback that helps improve the learning process. Song data such as note pitch, duration, and string and fret positions are collected to display notes on the virtual fretboard for the user to visualize learning concepts. This project expands on the use of MR technology in educational fields for a more immersive learning experience. Future work includes broadening this application to adapt other instruments taught by Catfish Alley Studio such as the ukulele or violin.
Ryan Anderson, Landon Baker, Chase Dowdy, Anousith Keomaly and Edgar Sanchez
Dr. Zakaria El-Awadi
CLEAR is a web planner app catered towards college students with predictive scheduling, AI consulting, and syllabus parsing. Planner apps with modern features, such as predictive scheduling, are often locked behind paywalls. College students often don’t have the budget to afford monthly subscriptions for planner apps. As a result, students are left with basic planner apps. Our project aims to give students a modern planner app.
To create this project, we used Ruby on Rails as the framework, Postgres for the database, and Ollama for the AI agent. A production branch of our project was also set up for user testing by college students. We received feedback and insights from testing which allowed us to further develop the app towards user needs. Users found our UI and calendar to be intuitive and easy to navigate. Additionally, users thought our syllabus parser was useful. From our findings, our project can be an intuitive tool for college students to plan out their course and personal schedules. Future steps would involve expanding on features and implementing a revenue stream.
Trey Harrelson, Michael Kenny, Austin Newsom and Jacob Pace
Hirsch Guitar Academy
Our project focuses on creating a web application that can eventually be integrated with a mobile application for Hirsch Guitar academy. Our sponsor requested that we need to address 2 major software hurdles for the website: we need to create a course management module similar to Canvas and a Facebook styled web forum akin to Facebook. Our group has split into 2 roles monitoring backend and frontend development of the website. Each sprint we conduct research and develop appropriate modules for the website.
Jacob France, William Herbert, Chris LeBlanc, Jake Touchet and Kyle Wilson
B4iGo
Roselin Beaulah, Principal Project Manager
Ramana Bandili, CEO
B4iGo is a secure, identity-backed vault designed to give individuals true ownership of their digital footprint. Currently, personal records are fragmented across platforms, leaving families locked out during emergencies. B4iGo solves this by securing digital artifacts with user-controlled cryptography and programmable access contracts, ensuring safe, revocable delegation of data. However, a current bottleneck in the B4iGo ecosystem is its reliance on manual data entry to populate vaults. This creates a persistent time tax, causing users to neglect recording vital information. Because a significant portion of actionable personal data flows through email, our project, the B4iGo Hybrid Authority AI Prototype, bridges this gap by automating vault curation. We engineered an AI-driven workflow that securely parses user email to identify and extract critical artifacts. To maintain B4iGo’s strict privacy standards, our system employs a "hybrid authority" (human-in-the-loop) model. The AI proposes additions, but explicit user authorization is required before data is committed. This eliminates manual entry friction while preserving absolute user control.
Hayden Brimage, Hassaan Bukhari, Timothy Farley, Nathan Gremillion
Dr. Pradeep Chowriappa
When planning out a garden’s future year in advance, it can often be difficult to visualize what one may want it to look like at a given time, and to then work backwards to determine when each plant must be seeded to result in the garden looking that way at that time. Our project, Leafy Ledger, seeks to alleviate this problem. Our project will allow the user to create a virtual garden space which mimics their own, select a time of year, and in a sense “paint” what they want their garden to look like at that point in time. The software will then work backwards and flag when each plant in the garden should be planted to get the results the user wants. Furthermore, the user will be able to scroll back or forward in time, allowing them to create multiple “phases” of the garden which they may interpolate between, adjusting the phases to avoid ever having a period of emptiness for the garden. As a stretch, we’re also shooting for adaptability to user-reported weather conditions, warning the user if the anticipated plants are not expected to grow in the given conditions.
Alex Becker, Journey Harris, Jonah Hoffman and Keahi Temple
Dr. Jerry Berg
PGEmu is an emulation frontend designed to make retro gaming emulation more user friendly than it currently is. A lot of emulator frontends (mainly Retroarch) are functional but not very user-friendly, and our project is attempting to change this by making one with a good user experience. After research into the subject, we decided upon building the frontend from the ground up, meaning we started with the bare essentials (you can see the games you have, as well as launch them through the program), and began building up from there. The end result is a simple and clean program that allows you to view all of your retro games in one place, complete with social features like seeing what your friends are playing. We intend to publish this application so that other like-minded individuals can enjoy the frontend that we have crafted.
Grant Cooper, Russell Kelly, Jayden Toussaint and Julia Wilson
Dr. Eric Kuo
Organizations often review hundreds of applicants for a single position, raising concerns about efficiency, fairness, and consistency. Resume LLM addresses this challenge by developing a local, open-source large language model pipeline that summarizes resumes without making selection decisions. The project focuses on structured, evidence-based resume processing aligned with personnel selection standards. The system uses three stages: segmentation of resumes into sentences, classifying those sentences into knowledge, skills, and abilities (KSAs), and generating task-based summaries such as Functional Job Analysis or Behaviorally Anchored Rating Scale statements. Personally Identifying Information is removed before processing to reduce bias, and outputs and model metadata are stored locally for evaluation and version tracking. The pipeline runs entirely on local hardware using open-source models to preserve confidentiality and control costs. By converting unstructured resumes into standardized task statements, Resume LLM improves transparency, scalability, and auditability in applicant review and supports future evaluation of model accuracy and fairness.
Fariza Abdurakhimova, Elora Browning, Caleb Cameron and John Gagnon
Josh Coriell
For modern day software development, developers tend to work together, but not collaboratively. This leads to issues such as merger conflicts and version control system delays. This project will address this issue by developing a live collaborative Integrated Development Environment (IDE), allowing for multiple users to all work simultaneously on a file and being able to see each other’s updates in real time. This project is using Electron for the frontend and Django for the backend. Our main challenge at the start of this project was to learn Django since none of us had experience with it. We have developed pages for multiple pages, setup a database on the backend, and have a cloud that we will be using to store the files. The most significant outcome for this project was to develop live editing between different users. For our future work, we would like to implement complete version control separate from github to give our app its own unique aspect.
Barry Dees, Gavin Dominique, Christian Hall and Michael Tannehill
Dr. Jerry Berg
MockUp is an open-source illustration and animation software which seeks to address the lack of free animation software which focus on the creation of animation layout. MockUp can be compared to other software like Clip Studio Paint and Krita but with a more pointed focus on the loose work of the layout man as he establishes the foundation of professional 2D animation. This software also offers a free alternative to the standard of costly month subscriptions established by competitors. MockUp was created in C++ utilizing several open-source libraries to enable key features such as rendering, user interface, and image imports and exports. Overall, MockUp focuses on a streamlined, understandable, experience allowing for easy use from the animator with little to no learning curve while not sacrificing key features like a comprehensive brush engine, onion skins, and more.
Emilio El-Zahr, Ray Favaza, Juwan Joseph, Remy Rains and Garrett Williams
Dr. Kevin Cherry
The MechaMaker project is an interactive robotics simulation game developed in Unreal Engine seeking to lower the financial and technical barriers to entry in robotics. Traditionally, robotics requires costly hardware and complex programming environments, severely limiting accessibility for beginners. MechaMaker aims to address this by gamifying typical robot configuration and procedural programming within a physics-based virtual environment. Players configure their robots using an in-game HUD to attach modular components, such as manipulators and lifts, and then program how/when their robot uses those components in the level through an in-game IDE. Each level introduces constrained challenges to promote critical thinking and problem solving, requiring users to strategically configure and program their robot to solve each challenge. Within the game’s IDE, the player can access a component function library for controlling the components currently attached to their robot, which can be called anywhere within a player’s program as they define their robot’s behavior. This program is coded through structured text using the Almost Procedural Programming Language (APPL) developed for the game, or a visual code block programming style inspired by MIT’s Scratch. The result is an accessible, engaging platform capable of teaching foundational robotics and programming concepts without prohibitive hardware costs.