Projects

Evolutio is a group of technology professionals convinced that business problems have significantly simpler solutions than the market is led to believe. These solutions span across the globe, including the non-profit Elephants, Rhinos, and People (ERP), a group founded to preserve and protect Southern Africa’s wild elephants and rhinos.

As part of their initiative to preserve and protect wildlife, ERP uses drones, or unmanned aerial vehicles (UAVs), to monitor elephants at the Rietvlei Reserve in South Africa.

Wildlife is threatened every day by not only poachers, but also by the destruction of food sources, the disruption of habitat by tourists and natural threats such as floods, wildfires, and drought. In a 400,000-acre park, it is impossible to detect and monitor threats without an automated system.

Our Conservation Threat Detection system serves two primary functions: auto-identify threats in drone footage and inform rangers of these identified threats in real time.

ERP pilots fly drones equipped with cameras throughout the reserve and our system automatically detects any threats, including cars, humans, fires, and floods, from the camera feed in real time.

If a threat is detected, nearby rangers are informed of the threat and its location through a graphical user interface (shown on the right), together with silent notifications conveyed through vibration motors mounted in our custom-designed ranger vest.

Our system allows ERP to monitor large areas of land in real time without the need for ERP personnel to manually analyze hundreds of hours of drone video footage. This allows ERP to more quickly respond to imminent threats.

Our threat detection is done using neural networks built with TensorFlow. All components of the system communicate through Ethernet protocol and the main system runs on a Jetson Nano.

Team Members (left to right) Qingyang Li, Drew Schineller, Maddie Jones, Logan McDonald

Ford Motor Company: Ford Augmented Reality Owner’s Manual

Ford Motor Company is a multinational automotive manufacturer headquartered in Dearborn, Michigan, employing 199,000 employees and producing a total of 5.9 million vehicles in the last recorded year. Ford designs and manufactures a full line of cars, trucks, SUVs and electric vehicles under both the Ford and Lincoln brands.

Every Ford vehicle comes with a printed owner’s manual containing more than 300 pages of basic information pertaining to the operation and maintenance of the vehicle. This manual is cumbersome, difficult to navigate, and has not evolved with the technology inside the vehicles.

Our Augmented Reality Owner’s Manual application provides an intuitive and accessible digital version of the owner’s manual with augmented reality (AR) capabilities.

The interior of the vehicle is displayed from the driver’s perspective using the phone’s camera. From this screen, interactive digital content is overlaid using AR, enabling users to quickly access resources.

When a user clicks on an interactive component of the augmented reality display, a list of relevant content is displayed. This content includes a digital version of the corresponding owner’s manual section, tutorial videos and answers to frequently asked questions. Alternatively, the same content is accessible through the search bar from the app’s homepage.
Authorized Ford employees create, edit and delete vehicles and manage any associated content through the web application. This content is accessed via the iOS app for the respective vehicle.

Our iOS application leverages Swift and ARKit to provide an AR experience. The web application is built using the ReactJS framework. The web application and iOS application are linked through an API and database hosted by Amazon Web Services.

Team Members (left to right) Shadman Rahman, Shawn Peerenboom, Torrin Bates, Ryan LaHaie, Jiahao Wang

General Motors: Open Source Intel

General Motors (GM) is a multinational automotive manufacturer headquartered in Detroit, Michigan. GM is ranked #13 on the Fortune 500 for total revenue and is the largest auto manufacturer headquartered in the United States.

Maintaining strong information security is a priority for GM to protect sensitive information that could compromise asset security and communication privacy. Publicly visible credentials grant unauthorized parties the opportunity to infiltrate GM assets and view private communication networks.
Our Open Source Intel system automates the discovery of security threats by collecting and analyzing information from various public code repositories on the internet such as GitHub, GitLab and Bitbucket.
Confidential intellectual property (IP) such as GM usernames, API keys and code snippets are displayed on a user-friendly web application. When a threat is discovered, relevant information about the IP leak is displayed so that GM employees can quickly act to mitigate the threat.
A machine learning service gives each discovered leak a confidence score. If a threat is assigned a high enough confidence score, employees are notified via text message and/or email.

Open Source Intel automates the currently manual process of discovering the warning signs of a leak and drastically increases employee effectiveness by letting them focus on threat mitigation instead of threat discovery.

The Python data collection pipeline is orchestrated using Celery, pipeline data is stored temporarily in Redis, and code is processed using PyDriller. A trained scikit-learn machine learning model quantifies each hit discovered by the pipeline. Open Source Intel stores data in a PostgreSQL database. This database then feeds the Python Django web application for display.

Team Members (left to right) Taylor Zachar, Ben Buscarino, Will Crecelius, Qiming Ren, Igli Ndoj

Harvard Law School: StackLife 2.0: Library Search and Display Tool

Located in Cambridge, Massachusetts, Harvard Law School is arguably the most prestigious law school in the world and is home to the world’s largest academic law library. The school’s faculty consists of more than 100 full-time professors and more than 150 visiting professors, educating students and delivering research on traditional and emerging legal fields.

The Harvard Library has consolidated approximately 15 million records describing documents from multiple sources surrounding Islamic policy, law and history. Access to this data is very limited with no user-friendly way of viewing it, depriving researchers of valuable information that could be beneficial to their research efforts.

Harvard Law School wants to facilitate universal access to this data in order to preserve a significant part of our shared world heritage, as well as promote new and data-centric research.

Our StackLife 2.0: Library Search and Display Tool consolidates the data for these resources and allows researchers to access it in an easy-to-use web application.

Our system allows researchers to filter records based on multiple search parameters in order to find the resources that are most relevant to them. It also enables users that are registered with the system to save search parameters to their profile and then build their own custom collection of resources.

Once a user finds a desired resource, our application provides the relevant data about that resource, including where the user can retrieve it. Our system also provides data visualization capabilities to enable researchers to plot data.

Our application is built using the Flask web microframework for Python with HTML, CSS, JavaScript, and Bootstrap. The data is stored inside of a relational database system using MySQL 8.0 that is hosted using Amazon Web Services.

Team Members (left to right) Allison Mutka, Brandon Field, Zian Gong, Jake DeNell

Herman Miller: Measuring Workspace Impact on Employee Experience

Herman Miller, a 100+-year-old company from Zeeland, Michigan, is an industry leader in home and office furniture. Known for its history of design innovation, Herman Miller dedicates research to office space quality in an effort to quantify the effectiveness of different workspace layouts. Currently, sensors are employed to measure utilization of workspace areas.
Sensor solutions, however, do not provide information regarding employee satisfaction, or sentiment, towards a specific workspace.

Our Measuring Workspace Impact on Employee Experience application allows employees to use kiosk stations and their mobile devices to input sentiment about specific workspaces. Additionally, our sentiment analysis web platform derives quick data insights from the survey results.

Users have the option to log in to the mobile application using their company code and either their user identification or continue as a guest. Registered users can view their rewards and statistics pages.

When a user with the mobile application leaves a workspace, a proximity beacon sends a notification prompting sentiment questions about the workspace. Kiosk stations are available in key workspace locations across a floor plan.

Collected data is displayed through the analytics web platform, accessible only by Herman Miller administrators. The platform allows users to better understand how to alter workspaces to increase employee satisfaction.

Our software uses Amazon Web Services including their relational database service, natural language processing, and API gateways for our back end to collect and analyze data. Estimote Proximity Beacons detect participating user locations on their Android or iOS devices. The web analytics platform is built using ReactJS and displays data using the Google Analytics API.

Team Members (left to right) Sophia Frankel, Jake Baum, Lynn Dai, John Nguyen-Tran, Marla Mae Defensor

Learning A-Z: Sandwich Builder Parts of Speech Guessing Game

Learning A-Z is an education technology company that expands students’ literacy through thoughtfully designed tools and resources, equipping students with the skills they need to succeed in the classroom.

Elementary school students use Learning A-Z’s software for multiple subjects and are familiar with the content and style. When a new resource is added, students spend less time learning the software and more time learning the material.

Our Sandwich Builder Parts of Speech Guessing Game provides a fun and interactive learning experience for students. The game is designed with Learning A-Z’s style and content, allowing the students to focus on learning the parts of speech of different words.

When a game is started, an empty outline of a sandwich with a part of speech in each layer is displayed in addition to a list of randomly chosen words.

If the student drags a word to the correct part of speech, the corresponding layer of the sandwich fills with a pleasant, correct color. If the student is incorrect, the layer of the sandwich fills with a mold-like color.

Once all parts of the sandwich are displayed, the student submits their work. If correct, the student is awarded 50 stars, the common currency for Learning A-Z software.

After correct completion of the sandwich, the user enters a bonus round, where they must correctly select a single word that matches two parts of speech. This round is worth an additional 10 stars.

Our Sandwich Builder Parts of Speech Guessing Game is developed using Angular for the front end of the web application and Swift for the iOS platform. It communicates with the MySQL database using PHP.

Team Members (left to right) Harry Mathon, Aarish Medhora, Raunak Shivkumar, Yibei Huang, Samantha Zielinski

Lockheed Martin Space: SmartSatTM Satellite App Store

Headquartered in Littleton, Colorado, Lockheed Martin Space is one of four business areas comprising Lockheed Martin, an American global aerospace, defense, security and technologies development company that employs 110,000 people worldwide. From the Orion spacecraft to satellites that can be reconfigured while in orbit, Lockheed Martin Space is a global leader of the space sector.

The Lockheed Martin Space SmartSatTM system introduces a universal software format that secures and standardizes satellite applications, allowing for frictionless collaboration on projects and compatibility across many different Lockheed Martin satellite models.

Our SmartSatTM App Store provides a web-based marketplace for browsing, uploading, and installing mission-ready applications directly to live satellites. Operators manage their entire fleet directly through our web page, enabling and disabling applications installed on satellites with the press of a button. Lockheed Martin Space and third-party satellite application developers alike are granted tools to collaborate efficiently on our platform, with shared access to projects and satellites.

Every new application uploaded to our app store is put through our automated compatibility testing to assess on which Lockheed Martin satellites the software can be deployed. This can save hundreds of hours of development by allowing a single piece of software to be deployed on multiple different satellite models.

Our web application stack consists of ReactJS UI components, Flask for back end, and a PostgreSQL database. The storage and distribution of SmartSatTM applications is done through Nexus, and compatibility testing is automated using a continuous integration server.

Team Members (left to right) Sailesh Gundepudi, Brian Fuessel, Peng Sun, Daniel Webb, Tony Miller

MaxCogito: Identity Based Communication and Content Services

Founded in 2019, MaxCogito provides software tools for customers to understand and work with their data.

Most companies are blind to the content of the messages passing through their servers. However, recent regulations such as the European Union’s General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) require enterprises to understand the data they store.

Our Identity Based Communication and Content Services platform automatically categorizes and analyzes every message that passes through a company’s servers to help them remain compliant with any regulations, and also to identify any valuable information that might otherwise have been missed.

Businesses using our product select the types of information to monitor, which can include data that could result in regulatory violations, contain trade secrets, or personally identify an individual.

As internal and external messages are sent to and from the employees of an enterprise, our platform automatically searches for content based on each company’s indicated needs. Any messages containing sensitive information are automatically tagged before being forwarded to the final recipient.

Our service generates intuitive reports on the content of the messages that come through their servers so that administrators and compliance officers can easily find and assess the quantity and type of sensitive data they currently have on their servers.

Our platform greatly simplifies the data analysis process, saving companies hundreds of man-hours while also avoiding fines for policy violations.

The Identity Based Communication and Content Services platform uses an SMTP server to collect messages, Apache Tika to extract text from each message, and Elasticsearch to index the data. Connecting the back-end services are RESTful APIs written in Java.

Team Members (left to right) Harrison Samoy, Bryan Hitchcock, Tengjiao Wang, Conor Sands, Andrew Kim

Meijer: Reducing Shoplifting Using Machine Learning

Founded in 1934, Meijer is the pioneer of the modern supercenter with 242 stores across the Midwest.

Every year, an estimated $10-30 million in assets are lost due to organized shoplifting. Meijer has identified behavior strongly associated with shoplifting, including short or long dwell times in high risk areas, leaving the store without passing through a point of sale, as well as leaving the store using employee or emergency exits.

However, Meijer stores do not have the manpower to watch and monitor every shopper who comes through their doors.

Our Reducing Shoplifting Using Machine Learning project automatically tracks Meijer shoppers throughout the store to identify suspicious behavior to prevent shoplifting.

Meijer has installed Mist wireless access points throughout their stores, which gives them the ability to track the general location of shoppers during their time in the store.

Our system tracks, in real time, the paths various shoppers take. It then uses machine learning to determine the probability that a given shopper is engaged in illegal shoplifting behavior.

If any suspicious activity is identified, the Meijer Asset Protection team receives an alert on their smartphone regarding the incident. The employee then uses that information to review the incident using the store surveillance system integrated into our desktop app. If an incident is confirmed to be shoplifting, the device number associated with the shoplifter is stored for future alerts.

Whenever a device that has previously engaged in shoplifting reenters a Meijer store, employees are notified and action can be taken to prevent further acts of shoplifting.

Our desktop app and mobile apps are written in C#. Our database is on Azure SQL. Our machine learning algorithm is written in Python and devices are tracked via Wi-Fi and Bluetooth using Mist access points.

Team Members (left to right) Xiaojun Wang, Billy Ochab, Justin Marinelli, Matt Schafer, Jesse Stricklin

Michael Sadler Foundation: Gamifying GameChang3rs

The mission of the Michael Sadler Foundation is to inspire and empower students in building their personal legacies. The foundation uses their Six Pillars of character as stepping-stones for this growth and does so with the GameChang3rs Program.

GameChang3rs is a program that provides K-8 students with tools to help them develop strong character, make good choices, and mature both socially and emotionally. GameChang3rs ambassadors are volunteer high school students who teach and mentor elementary school students.

Students and ambassadors meet once a month during the school year. However, between these meetings, the Michael Sadler Foundation is concerned that their students do not retain the important information from the GameChang3rs lessons.

Our Gamifying GameChang3rs project is a web-based platform designed to keep students continually engaged with the Six Pillars material throughout the time between meetings.

Gamifying GameChang3rs contains a variety of educational games designed to teach K-8 students lessons about the Six Pillars of character. Students have fun and earn points all while interacting with material in a fun and educational manner.

GameChang3rs administrators can view statistics about which games are the most popular, how many games are being played a day, and how many students are logging in to the system. To adhere to privacy regulations, no information that can identify students is stored. Instead, this information is used to identify which games are effective to guide the development of future games.

The front end of our system is built using Angular 8, while the back end is implemented using Express. In addition, the Unity game engine, along with WebGL provides a simple solution to develop native web games. Game and user information is stored in a MongoDB database.

Team Members (left to right) Matthew Vedua, Daniel Marinetti, Tristan Özkan, Dima Zhang, Lina Jebara

Michigan State University CSE: Using Sensors to Study Human Behavior

The nation’s pioneer land-grant university, Michigan State University (MSU) is home to nationally ranked and recognized academic, residential college, and service- learning programs.

Among the fastest-growing academic programs at MSU, the Department of Computer Science and Engineering (CSE) hosts nine research laboratories and equips students with practical skills that enable them to adapt to changing technology.

Dr. Mohammad Ghassemi is an associate professor in the computer science department at MSU. Dr. Ghassemi researches human health and behavior using machine learning, and is the director of a study at MSU in which the behavior and interactions between small groups of individuals are examined.

Our Using Sensors to Study Human Behavior system transforms the laboratory dedicated to this study into a “smart” meeting space. Human movement is captured using cameras, dialogue is collected using microphones, and an electroencephalogram (EEG) is used to study brain activity.

The collected data is used to train machine learning algorithms, detect anomalies in human conversation, and track eye movements using strategically mounted cameras.

The data is streamed to our website (called The Data Hub) where it is viewed and analyzed by researchers and stored for later research and analysis. Additionally, The Data Hub allows researchers to set event triggers in response to data. Triggers can be as simple as a text notification or as complex as a change in the environment of the lab.
The lab contains an EEG as well as multiple cameras and a microphone connected to Raspberry Pis that stream data to our remote server running MySQL. The Data Hub is written in Python using the Flask web framework.

Team Members (left to right) Lianghao Shu, Ben Seeger, Rainier Devolder, Taylor Whitacre, Merryn Marderosian

Michigan State University ITS: Degree Navigator

Michigan State University is a public research institution founded in 1855. The Information Technology Services (ITS) unit delivers and maintains effective technology resources for students, faculty and staff.
There are 97 majors of study and more than 100 minors available to students. While only required to complete one major, students can complete any combination of majors and minors as well as participate in the Honors College. Each of these programs has a unique set of graduation requirements, causing students to struggle to keep track of which have been met and which still need to be completed.

Our Degree Navigator application provides an easy-to-use interface for students to check their progress in each of their chosen programs. The landing page displays a summary of each program in which the student is participating – a major, a minor, or the Honors College.

Clicking on any of these programs navigates the user to a list of program requirements in the form of either a specific course or a list of courses. Incomplete requirements are listed at the top. The user can expand a requirement and get more detailed information about which classes can fulfill that requirement or can expand all requirements to view a detailed description of the program.

In addition, students can view recommended four-year course schedules for each major provided by Michigan State. Each course is accompanied by a symbol that represents whether the course was already taken, currently being taken, or not yet taken by the student.

Degree Navigator is developed with Swift for iOS, Kotlin for Android, and ReactJS for web. It uses an AWS API to access information stored in a DynamoDB database via Lambda functions written in Python.

Team Members (left to right) Tony Fedewa, Chad Capuzzi, Christian Velkovich, Sarah Johanknecht

Headquartered in Seattle, Amazon is the world’s largest online retailer and is also the world’s largest cloud services provider with their Amazon Web Services (AWS) products.
As a leader in the technology sector, Amazon has access to massive amounts of data. They employ teams of data scientists to analyze this data to improve Amazon’s various offerings, including their product recommendations.
The task of finding the best dataset for a problem is time- consuming and requires significant manual work, including looking through thousands of individual files that are stored in many different locations. This process takes up a substantial amount of time that could be better used for development.
Our Amazon Data Hub software streamlines dataset acquisition with an easy-to-use website that allows data scientists to automatically search through Amazon’s collection of data.
When an Amazon data scientist uploads a dataset to our Amazon Data Hub repository, it undergoes automated analysis. This includes object detection and speech recognition for images, videos and audio, as well as statistical analysis of numerical data.
Data scientists use the web application to search through our catalog of datasets. Search results include information provided when the dataset was uploaded, as well as information from our automated analysis. Intuitive visualizations of each dataset allow users to quickly evaluate the relevance of each dataset.
The Amazon Data Hub decreases the time it takes to find suitable datasets from hours to minutes, allowing data scientists to spend their time on more important work.
Our system uses AWS’s scalable products, including S3, DynamoDB, Rekognition, Transcribe, Lambda, Elastic MapReduce, and Elasticsearch, to store, process and search the datasets. Python Flask is used to connect our back end with our ReactJS front end.

Team Members (left to right) Robert Ramirez, Cameron Nejman,Josh Barnett, Austin Cozzo, Dan Farat

Amazon: Amazon Data Hub

Headquartered in Seattle, Amazon is the world’s largest online retailer and is also the world’s largest cloud services provider with their Amazon Web Services (AWS) products.As a leader in the technology sector, Amazon has access to massive amounts of data. They employ teams of data scientists to analyze this data to improve Amazon’s various offerings, including their product recommendations.The task of finding the best dataset for a problem is time- consuming and requires significant manual work, including looking through thousands of individual files that are stored in many different locations. This process takes up a substantial amount of time that could be better used for development.Our Amazon Data Hub software streamlines dataset acquisition with an easy-to-use website that allows data scientists to automatically search through Amazon’s collection of data.When an Amazon data scientist uploads a dataset to our Amazon Data Hub repository, it undergoes automated analysis. This includes object detection and speech recognition for images, videos and audio, as well as statistical analysis of numerical data.Data scientists use the web application to search through our catalog of datasets. Search results include information provided when the dataset was uploaded, as well as information from our automated analysis. Intuitive visualizations of each dataset allow users to quickly evaluate the relevance of each dataset.The Amazon Data Hub decreases the time it takes to find suitable datasets from hours to minutes, allowing data scientists to spend their time on more important work.Our system uses AWS’s scalable products, including S3, DynamoDB, Rekognition, Transcribe, Lambda, Elastic MapReduce, and Elasticsearch, to store, process and search the datasets. Python Flask is used to connect our back end with our ReactJS front end.

Team Members (left to right) Robert Ramirez, Cameron Nejman,Josh Barnett, Austin Cozzo, Dan Farat

Amazon: Amazon Data Hub

Headquartered in Seattle, Amazon is the world’s largest online retailer and is also the world’s largest cloud services provider with their Amazon Web Services (AWS) products.As a leader in the technology sector, Amazon has access to massive amounts of data. They employ teams of data scientists to analyze this data to improve Amazon’s various offerings, including their product recommendations.The task of finding the best dataset for a problem is time- consuming and requires significant manual work, including looking through thousands of individual files that are stored in many different locations. This process takes up a substantial amount of time that could be better used for development.Our Amazon Data Hub software streamlines dataset acquisition with an easy-to-use website that allows data scientists to automatically search through Amazon’s collection of data.When an Amazon data scientist uploads a dataset to our Amazon Data Hub repository, it undergoes automated analysis. This includes object detection and speech recognition for images, videos and audio, as well as statistical analysis of numerical data.Data scientists use the web application to search through our catalog of datasets. Search results include information provided when the dataset was uploaded, as well as information from our automated analysis. Intuitive visualizations of each dataset allow users to quickly evaluate the relevance of each dataset.The Amazon Data Hub decreases the time it takes to find suitable datasets from hours to minutes, allowing data scientists to spend their time on more important work.Our system uses AWS’s scalable products, including S3, DynamoDB, Rekognition, Transcribe, Lambda, Elastic MapReduce, and Elasticsearch, to store, process and search the datasets. Python Flask is used to connect our back end with our ReactJS front end.

Team Members (left to right) Robert Ramirez, Cameron Nejman,Josh Barnett, Austin Cozzo, Dan Farat

Headquartered in Sunnyvale, California, Proofpoint is a cybersecurity firm focusing on enterprise-level threat tracking, mitigation, and elimination. While Proofpoint is known for client endpoint protection, they also employ an extensive R&D infrastructure for handling and analyzing new malware.

Analyzing malware is challenging. Viruses, spyware, ransomware and other malicious programs come in many forms. To protect its customers, Proofpoint analyzes malware using tools called sandboxes, which are isolated computing environments where malware can be tested safely. The industry standard is a short-term analysis on malware samples for 2 to 15 minutes each.

However, malware developers know of sandboxes and often design their malware to change its behavior weeks or months after infecting a system. Because of this, short-term malware analysis is not always effective in determining the effects of certain malware.

Our Predictive Engine for Long-Term Malware Detonation platform offers an intuitive web dashboard to efficiently manage malware samples and analysis, as well as a service to quickly identify unique and duplicated malware samples.

Our website allows Proofpoint analysts to upload malware samples, view the results of previously analyzed samples, monitor currently running malware, and view overall system statistics.

When a malware sample is uploaded from our dashboard, it is automatically analyzed in a few minutes to determine if it is unique or similar to previously run samples. Because running the sandbox for long periods of time is expensive, our system will prioritize unique malware samples for long- term analysis and discard duplicated samples to save on processing time and money.

Our predictive engine is implemented in Python, using Cuckoo, YARA, and OPNsense. Our web app uses Angular 8 for the front end, and Python Flask and MongoDB for the back end.

Team Members (left to right) Izzy Dove, Joshua Wilson, Alex Kendall, Geoffrey Witherington-Perkins, Samuel Gendelman

Technology Services Group: Volunteer Onboarding and Patient Visit Management

Founded in 1996 in Chicago, Technology Services Group (TSG) is an expert in data and document management. TSG has many clients across a wide range of industries and is a leading provider of content management solutions.

TSG also works with nonprofit companies in the medical industry who are often burdened with large volumes of documents and forms. Two of the most common include employee onboarding documentation and patient forms. Human resource representatives spend much of their valuable time tracking and collecting these documents.

Our Volunteer Onboarding and Patient Visit Management system integrates the power of the Microsoft Azure cloud with TSG’s existing software, OpenContent Management Suite (OCMS) to create new OCMS dashboards to streamline the creation, upload, and management of employee onboarding documentation and patient forms.

Our newly designed web dashboards simplify the employee onboarding process by aggregating the completed and outstanding necessary documentation to one central location. Employees can use our visualization tools to quickly track the progress of a new employee’s onboarding and monthly patient visits.

Additionally, employees can automatically record, save, and track patient visits without the need for paper forms. Employees can search, add, and update any required patient documents quickly and easily.

Our system saves employees significant time and energy. Patient tracking is now automatically handled in one convenient location that can be accessed by anyone at any location.

Our Volunteer Onboarding and Patient Visit Management system utilizes Apache Tomcat, HTML, Java, JavaScript, and the Microsoft Azure cloud service HDInsight.

Team Members (left to right) Lazaro Cruz, Will Giger, Xiaokuan Zhang, Genya Dobrev, Jacob Harris

TechSmith: Smart Camera

To help its customers communicate more effectively, TechSmith assists in the creation of images and videos. Its flagship products, Snagit and Camtasia, are used by more than 30 million customers, worldwide.

Many customers of TechSmith do not have a background in video production. This lack of experience can often lead to less than professional content.

Our Smart Camera software assists TechSmith users in creating better video content through intuitive, easy-to-use mobile and web applications. The content created using our system can be easily used with TechSmith’s video editing software, Camtasia.

The Smart Camera iOS application offers a suite of tools to give video creators feedback and advice on filming in real time. As the user films a video, our software automatically analyzes video frames continuously to provide feedback on the lighting quality, as well as the framing of their video scenes. Example feedback can be seen in our screenshots on the right.

Smart Camera also supports a live teleprompter feature on mobile devices to display prepared scripts during filming. The teleprompter non-obtrusively overlays the user’s script on the camera view (shown on the right).

The Smart Camera web dashboard allows users to create and manage their scripts, which can be exported to the Smart Camera mobile application for filming. Additionally, the web dashboard aggregates all completed video assets, including scripts and raw video files. These assets can be exported automatically from the Smart Camera mobile application.

Our web application is made using JavaScript, C# and the ASP.NET core framework. Our web application, video storage, and database are hosted on Microsoft Azure. The mobile application is written in Swift.

Team Members (left to right) Omo Irumundomon, Zhaolin Liu, Nathan Anthony, Drew Bensinger, Amy Kim

United Airlines: Safety Reporting and QC Audit Center Mobile App

United Airlines is the world’s second largest airline, operating approximately 4,900 flights a day and transporting over 150 million passengers a year out of 362 airports around the globe. To maintain its fleet of 1,300 aircraft and ensure successful flights, it is crucial to identify and resolve safety concerns and hazards.

Over 88,000 employees work for United Airlines, yet less than 10% of the airport operations employees fill out safety reports, called GSAP forms. This is because GSAP form submission can only be done on a desktop computer, meaning that employees in the field must go inside and find an available computer in order to fill out a form.

Similarly, the Quality Control (QC) Audit forms are currently filled out with pen and paper while the employee is in the field, and then require additional time for the employee to input that information onto a computer.

Using our Safety Reporting and QC Audit Center Mobile App, employees can efficiently and effectively fill out GSAP and QC Audit forms in the field.

Within the application, users can create, save and submit forms. Saved forms can be accessed at a later time via desktop or mobile app, and are available until the employee submits the form or the form expires. The application caters to an employee’s position and only shows the types of forms that are applicable to their specific role.

This application helps employees save time and effort by enabling the completion of GSAP and QC Audit forms in the field. The ease of use incentivizes more employees to fill out safety reports, increasing participation in the GSAP program.

Our Safety Reporting and QC Audit Center Mobile App is built with Swift and Texture for iOS, Kotlin for Android, Python and Django for the API, and a Microsoft SQL Server database.

Team Members (left to right) Camille Emig, Ivan Zhang, Allison Lollo, Josh Jarvis, Tudor Robaciu

United Airlines: Virtual Reality Aircraft Walkaround

United Airlines is one of the world’s largest airline companies. Headquartered in Chicago, they operate 4,900 flights a day from 362 airports worldwide.
Prior to each of these flights, technicians perform an aircraft walkaround to identify potential defects and issues with the aircraft. Training for this task is mainly done on the job, with little control over the types of defects or issues that can be demonstrated.

Our Virtual Reality Aircraft Walkaround software mitigates this problem by allowing technicians to be trained through virtual reality simulations on iPads and Oculus Quest headsets.

Virtual Reality Aircraft Walkaround features both user training and testing modes. In training mode, the technician performs a guided walkaround in virtual reality. Defects appear on the aircraft with popup boxes that display information about what the technician should look for in that location.

In testing mode, the technician spots and marks defects in an unassisted walkaround. After the technician completes the walkaround, a report that provides a summary of the technician’s performance is generated. The report is saved for instructors to assess a user’s progress. The technician is given the opportunity to return to the aircraft and review any mistakes.

A variety of aircraft models are available for both training mode and testing mode walkarounds. Each aircraft has a number of preset scenarios with predetermined defects. In testing mode, the technician may also choose a randomized scenario that spawns different defects on the aircraft each time.

Our software allows United Airlines to train technicians quickly, effectively, and cheaply.

Our software is developed with the Unity game engine. Scripts are written in C#, and the reports are saved using Unity Analytics. The software is available on both the iPad and the Oculus Quest.

Team Members (left to right) Caitlin Brown, Ellie Locatis, Jiachen Lin, Jacob Turcano, Cheney Wang

United Airlines: Training Scheduling and Optimization System III

United Airlines is a major international air-carrier operating 4,900 flights per day from 362 airports. Operating an airline requires diligence in all logistical and technical aspects to ensure the proper flight experience for “Every customer. Every flight. Every day.”

Within United Airlines, the TechOps training division is responsible for the operations of aircraft and their important maintenance. The TechOps team leverages a sub-team of 45 instructors to teach a catalog of 100+ courses for around 700 classes per year to their skilled team of 7,000 technical staff members. Currently, the orchestration of scheduling these courses is the responsibility of a single individual.
Our Training Scheduling and Optimization System III provides a production-ready web app to facilitate United’s maintenance training schedulers to schedule instructors, students, classrooms and courses across the country.

A scheduler uses our mobile compatible website to add classes to the schedule manually or make use of the schedule optimizer. The schedule optimizer automates the scheduling of multiple classes at a time. Our optimizer can schedule months of classes in a few minutes, compared to the many hours it currently takes to schedule these courses.
An automatic email system alerts the scheduler of important changes, including when new training requests or employee availability changes arrive.

Our platform streamlines the scheduling process for United Airlines’ TechOps division, allowing their employees to spend their time on more important tasks.

Our software is built using ASP.NET Core 3.1, Angular 8, Node.js, an Entity Framework, and an Azure SQL Database. The web app is hosted on Azure Cloud Platform.

Team Members (left to right) Sam Vitu, Weiyu Huang, George Poliakov, Sean Wisely, Amanda Hackbardt

Urban Science: AutoHook Mobile Redemption Tool

Headquartered in Detroit, Urban Science is internationally renowned for providing data-driven, science-based solutions to problems in the automotive, health, and retail industries. AutoHook is a subsidiary of Urban Science and assists automotive dealers and OEMs in increasing walk-in customer traffic.

AutoHook currently provides a voucher redemption service used by auto dealerships. This service allows dealers to redeem customer’s vouchers at the dealership for rewards, such as gift cards. Currently, the redemption service is solely available as a desktop website, meaning that all redemptions must be performed using a computer.

Our AutoHook Mobile Redemption Tool enables auto dealers to utilize any mobile device with a web browser to redeem vouchers for their customers.
Dealers have access to our redemption tool from anywhere, including the vehicle lot or during customer test drives. Additionally, our redemption tool utilizes the camera found in most mobile devices as a barcode scanner. This allows dealers to scan the barcode located on a customer’s voucher instead of being required to manually enter the relevant information.

Additionally, our AutoHook Mobile Redemption Tool offers intuitive visualizations and graphs of helpful statistics and calculated metrics to enable dealers to understand the effectiveness of their voucher campaigns. Also included in our platform is a dedicated question and answer section, which includes useful documentation and training videos.

AutoHook Mobile Redemption Tool is an online web application designed to work on all modern mobile browsers. The front end, created with Angular 8, is easy to extend and modify. The tool makes use of a supporting ASP.NET back end on 4.8 .NET Framework.

Team Members (left to right) Heng Yan, Devin Hook, Justin Perry, Torel Welsh

Vectorform: Rumble Test Suite

Vectorform, headquartered in Royal Oak, Michigan, invents digital products and experiences both for their own products and for the world’s leading brands.

Our Rumble Test Suite includes a Rumble device, an iOS application, and a web application. The Rumble Test Suite upgrades washing machines by recognizing, in real time, when a washing machine has finished running.

The Rumble device contains a sensor that detects the vibrations coming from a washing machine. Additionally, the Rumble device contains communication technology to wirelessly transmit the information from its sensors across the internet.

Using the data collected from the Rumble device, our Rumble Test Suite distinguishes when a washing machine is operating, and when it has finished. Our solution uses deep learning to make predictions on the state of the washing machine. Additionally, our solution is generalizable to any type of washing machine and wash cycle. This allows our platform to be widely deployed without any additional development overhead.

The iOS application alerts users when their laundry finishes, affording people more freedom to do other activities without worry of forgetting their laundry.

The iOS app configures the Rumble device while it is deployed, allowing developers to quickly diagnose and fix any problems. Our web app displays all wash cycle data from the Rumble device for analysis by Vectorform employees.

The Rumble Device contains an ESP32 and an accelerometer. The ESP32 is connected to an iOS device using Bluetooth Low Energy. The accelerometer reads vibration data that is pushed to our MySQL database over MQTT. The web application is implemented using HTML, CSS and the ReactJS extension Victory React for data visualization.

Team Members (left to right) Hyeungsuk Kim, Andreas Frame, Reis Wiedemann, Anna Quenon