Eight teams, including many Drexel University students and alumni, took home awards at the ninth annual Philly Codefest, presented by Comcast Cable and Drexel’s College of Computing & Informatics (CCI).
This year’s theme focused exclusively on developing novel hardware and software solutions to ensure AI’s safety and positive impacts on society, including challenge areas such as fintech, cryptocurrency & blockchain; cybersecurity; bias and ethical implications of AI; healthcare; accessibility; and environmental impact.
This year’s Philly Codefest was offered in a hybrid format, with teams forming and collaborating virtually from April 4 to 8, culminating in an in-person event on April 9 at the Quorum at 3675 Market Street.
The in-person event kicked off with a keynote speech from Comcast’s Director of Machine Learning, Rama Mahajanam, to more than 175 attendees, followed by in-person demo and judging sessions, a gaming tournament hosted by the Drexel Gaming Association, sponsor booths, and a Comcast Lounge. The April 9 event also featured talks by Ciright Founder and Chairman of the Board, Joe Callahan (BS engineering, information systems technology & marketing), and computer science PhD student and Co-founder and CEO of Lula, Adit Gupta.
In addition to presenting sponsor Comcast Cable, Philly Codefest 2022 was made possible by SICK Inc., SAP, Vanguard, Ciright, Urban Technology Project, CircleCI, Teachers&, Lula and Linode.
More than 25 hardware and software applications were submitted for judging on April 9. Codefest’s expert panel of judges – comprised of Drexel faculty, alumni, CCI Corporate Partners, sponsors and tech community leaders – selected the following projects as winners of Philly Codefest 2022:
Team Hygia (Philly Codefest 2022 Student Team Award)
Members: Abu Zaid Khan (BS computer science), Anomitro Paul (BS computer science), Kiruthik Sai (BS computer science), Hasan Shameer Muhammed (BS computer engineering), Hruday Vairagade (BS computer science), Akhil Shameer Mohammed (BS computer engineering)
What it is: An app to help substance abuse patients achieve a dependence-free lifestyle.
Inspiration: “Philadelphia is facing the greatest public health crisis in a century. Every neighborhood in the city is being hit hard by an epidemic of opioid use and overdose. Across all racial and ethnic groups, the number of deaths from drug overdose is higher than the number of deaths from homicide. Fatal overdoses in the first six months of 2021 rose by nearly 10% compared to the same time period last year, putting Philadelphia on track to be the center of the crisis. As students, in this city of brotherly love we felt the need to device a solution using our skillset to address the problem.”
What it does: “You’re in recovery from opioid addiction, and your walk to work takes you down the same streets where you used to buy heroin. The drug’s calling to you, still. Just then, your phone buzzes with a message that reads like a text from an old friend: ‘Hey, I know you’re near a risky area. You can do this.’ It’s from a notification from Hygia. It is one of the many features that the app has to help the user recover from their addiction. The app uses machine learning techniques to give the user an alert for the next month on how susceptible the user is to an overdose. It has an interactive map of resources like nearby buildings with Narcan, pharmacies, and hospitals that the user can use in times of need. It also has a medicine logger that one can use to upload their daily doses of medicine, and it will remind them when it's time to take it. Hygia also curates articles that make you feel good! Our Web crawler scans through hundreds of health-related news articles and provides the user with only the ones that are helpful and have a positive sentiment.”
Watch the Demo on YouTube
Team IntelDragon (Philly Codefest 2022 Collaborative Team Award)
Members: Christine Palmer (MS materials engineering and mechanical engineering ’16), David Freiberg (PhD candidate, materials science and engineering)
What it is: “A scalable artificial intelligence platform that identifies relevant cyber threats to an organization, industry, company, or product based on a written description of the topic of interest.”
Christine: “As this article puts it: ‘Cybersecurity teams sometimes see threat intelligence as the quick fix that will protect them from hackers and scammers. This expectation is largely overinflated. The fact is, there is no such thing as one-size-fits-all threat intelligence...A financial services company, for example, probably wants to pay close attention to website forgery and malicious contact forms aimed at deceiving targets into revealing their credit card and bank account numbers.’ I work in cybersecurity, and all the feeds out there currently contain a lot of information that is not relevant. Do a Google News search for "cybersecurity" and you'll see what I mean; as of an hour before the deadline (I'm a Drexel alumna, doing everything an hour before the deadline is normal), the top headlines were:
- Nike is Opening a ‘Technology Center’ in Atlanta Next Year
- Healthcare to Increase Tech Spending for Cybersecurity, Marketing, and AI
- Government Officials: AI Threat Detection Still Needs Humans
- Coro Banks $60 Million to Bring AI-based Cybersecurity to SMBs
- FDA Ramps Up Cybersecurity Efforts with Stricter Guidance for Devicemakers
Each of those headlines is potentially relevant to you, depending on who you are and what you're doing. If you're a small business owner in Atlanta, Nike opening a tech center might mean more business or a change in traffic patterns. If you're an investor, you might reevaluate Coro based on its latest purchase. If you're in the medical field, you need to know what new FDA regulations are coming down the pike. None of these articles are bad. But there are very few people for whom all five of those headlines are immediately relevant and useful. I think we can do better.”
David: “Unlike Christine, I'm not a cybersecurity professional, and everything I know about cybersecurity is what I've picked up through osmosis from having friends in the field and seeing the occasional relevant article. Which is to say, I know just enough to know how huge the field is, and not much more, and given that I don't even own a smartphone, my personal need for up-do-date information is probably less than most people in the tech field. However, I have worked with word vectorization quite a bit, and when Christine pitched the idea of using it to better understand cybersecurity, it was far too good an idea to pass up.”
What it Does: “Put simply, it attempts to solve the problem of too much information in the news by filtering out irrelevant news. Type in a brief description of what you want, or what you do, and it returns the articles which are semantically closest to your description, not just articles which happen to share the same keywords.”
Watch the Demo on YouTube
Team GAT: Gestural Artificial-Intelligence Technology (Freshmen Challenge; Comcast Cable Challenge – Build a Gestural Interface for Web Design)
Members: Ryan Brosius (BS computer science), Ebubechukwu Enwerem (BS computer science), Dalu Okonkwo (BS computer science)
What it is: “Do you want to communicate using gestures, are you looking for a new web experience? We GAT you. With GAT, you can communicate signs and gestures using just your webcam. No fancy tech needed.”
Inspiration: “The implementation of Artificial Intelligence (AI) in our daily lives is a trend that is growing significantly. AI technologies that can recognize and perform actions based on human movements are becoming more common. In the aftermath of a global pandemic, communication-using technology is on the rise. The state of affairs in AI and human communication using technology inspired us to create an AI model that can identify human movements and interact with a software interface based on the specific movements.”
What it does: “GAT: Gestural AI Technology is a collection of two AI models that can recognize human movements using a camera and perform specific actions based on specific human movements. It is divided into two models, one that is completely optimized in the browser, and one that can be run on your computer as an app. With these models, developers can map hand gestures to specific actions.
First Model (Website Optimized): Our First Model is an AI model that can be built into any website that recognizes human gestures using a camera and performs actions based on these gestures. We built the UI interface of a social media app, and assigned hand gestures to implement scrolling, mouse pointer movement, liking posts and automated typing of certain words.
Second Model (Computer System Optimized): Our Second Model is an AI model that runs on a computer system and recognizes human gestures using a camera and performs actions based on these actions. We trained our model to recognize hand gestures and use them to control the mouse pointer, click action, automated typing of certain words, backspace key, and the return/enter key. We built a messaging app to demonstrate the Web experience associated with this model.”
Watch the Demo on YouTube
Team Autonomous Ocean Cleaner (SAP Most Innovative Hack)
Members: Jordan Irgang (BS mechanical engineering), Sandeep Sharma (BS mechanical engineering), Allie Do (BS computer science), Victoria Young (BS computer science)
What it is: an AI-powered ocean cleaning boat.
Inspiration: “Every year, an estimated 8 million metric tons of plastic — about the size of nearly 57,000 blue whales - enters our oceans. Ocean pollution causes the death of one million sea birds and over 100,000 marine animals from plastic entanglement and ingestion per year. In addition to the deterioration of marine life, ocean contamination affects our water sources and the air we breathe — it supplies freshwater and nearly all of Earth’s oxygen. The ocean also moderates the climate, influences the weather, and affects human health. In other words, the ocean and humans are inextricably intertwined. According to rubicon.com, for every square mile of ocean, there are more than 46,000 pieces of plastic floating in it. Our team is inspired to improve the current polluted state of the ocean by building an autonomous boat (the Sundoo) that can detect, track, and pick up trash.”
What it does: “The Sundoo detects trash by using computer vision to pinpoint a floating object (or objects) and machine learning model pre-trained with the TACO Trash dataset to classify it as garbage. The coordinates of the object (restricted by the bounding box) are then passed as parameters to lock the object in for tracking purposes. As the boat approaches the tracked object, about 4 inches from the camera, the gripper captures the object and releases it into a trash bag or a designated trash area.”
Watch the Demo on YouTube
Team Discord Friend Finder Bot (The Future of Online Learning Challenge Winner)
Members: Evan Rusnak (BS computer science), Yashodha Ravichandran (BS computer science)
What it is: A Discord bot that utilizes existing Drexel classrooms to help people find possible friends in online settings.
Inspiration: “Due to the pandemic, we've all probably had online classes, and we all know they are not that fun. For us, the worst part was meeting people. So, when we saw this challenge, we immediately wanted to tackle how to solve this problem.”
What it does: “With this Discord bot you can add previous classes, goals for the current class, and personal hobbies. Then, when the teacher is ready, they can tell the Discord bot to pair students together based on these interests. Everyone will have a unique pair for the class, someone they can reach out to with questions and hopefully have a connection with.”
Team Gesture Interface and Online Experience (Best Diversity, Equity & Inclusion [DEI] Hack)
Members: Jay Lee (MS computer science), Shirley Qian (MS human-computer interaction & user experience)
What it is: a Virtual Reality Controller.
Inspiration: “Emerging complex new technologies and new functions increase memory loads of remembering the interactions or button locations on a website. Gesture interface design based on real world experience can make the learning process easier.”
What it does: “We designed a set of gesture interactions based on people’s experience from the real world. We expect the gesture interface we designed can reduce the memory loads and increase the user experience in people’s Web browsing activities.”
Watch the Demo on YouTube
Team altruiSMS (Best Hack for Social Good)
Members: Trinity Kleckner, Wahub Ahmed, Shi Jie Samuel Tan (all students from Haverford College)
What it is: “Phil is an SMS chat bot to notify the homeless and hungry of shelters and welfare distribution anywhere, at any time.”
Inspiration: “In Philadelphia, many homeless and hungry people do not have stable internet access. Hence, it is difficult for them to reliably communicate with them about where the distribution of food and daily necessities are taking place. Hence, we want to create a SMS chat bot that connects the homeless and hungry with voluntary welfare organizations and churches.”
What it does: “In a world where event communication is almost entirely online, we cannot omit those without stable internet, especially when it comes to events distributing the resources they need the most. Our SMS welfare-distribution notification system provides those who need it with not just alerts, but an SMS chat bot that is ready to assist them at any time! Through a conversation with Phil, our SMS chat bot, users can opt into receiving text notifications when local organizations are holding distribution events for items of their choosing (food, diapers, sanitary products, etc.). Organizations can register on our application to post upcoming events they are holding. They can also see when and where other organizations are distributing supplies to better coordinate efforts for effective distribution of daily necessities throughout the year. When registering, organizations are strongly encouraged to note if they are a 24-hour shelter. This information allows our chatbot feature to find the nearest shelter for anyone at any time. In accordance with our goal of making this chatbot as accessible and inclusive as possible, we ask for no personal information; we only communicate based on mobile messaging and request for the user’s items of interest, and the user’s location via a nearby street intersection. User information can be easily updated or deleted to protect the user’s privacy.”
Watch the Demo on YouTube
Team RunRouteFun (Most Creative Hack)
Members: Kaung Zan (BS computer science), Alim Imanmalik (BS computer science), Bao Dang (BS mathematics), Adam Terhaerdt (BS software engineering)
What it is: “An awesome Web app that helps you find creative routes for your next run.”
Inspiration: “We developed an application for both beginner and advanced runners to continue to stay motivated to run! [Team member] Bao started running last month and he quickly discovered that getting myself up and out of bed is the hardest part. Also, he found that running the same routes every day can quickly get boring which can lead to demotivation. While starting can be the hardest part, you don’t have to do it alone. Our application, RunRouteFun, is an awesome Web-based platform that helps you find creative routes for your next run.”
What it does: “Route generating. While preparing for a run can be the hardest part, you don’t have to do it alone. RunRouteFun will be your best running buddy who makes sure every run is a new challenge that is both athletic and artistic. Although further development is needed to improve the route recommendation, fun is guaranteed!”
Watch the Demo on YouTube