Hack AI 2024 Recap
The Ohio State University’s Artificial Intelligence Hackathon
This past weekend, I had the privilege of mentoring and judging at Hack AI, an artificial intelligence hackathon hosted by the AI Club at The Ohio State University. As a representative of 99P Labs, one of the event’s sponsors, I provided guidance to teams of students as they developed creative AI projects.
The atmosphere was incredibly positive and fostered a collaborative spirit throughout. Participants ranged from undergrads to graduate students, creating a diverse mix of talent and enthusiasm. Notably, many were just starting their tech journey. Their eagerness to gain hands-on skills made for a refreshing dynamic. This diversity of experience enriched the learning opportunities and highlighted the inclusive nature of Hack AI. The AI Club hosted two hands-on workshops during the event — one focused on Python and another on React — providing valuable skills development for attendees. The commitment from all involved showed the power of tech and education in shaping AI’s future.
Being part of this community is rewarding. 99P Labs’ involvement goes beyond sponsorship — it’s an investment in the future of tech and education. Mentoring and judging let me see our impact firsthand, as we help cultivate an ecosystem where curiosity, innovation and collaboration thrive. Engaging with members eager to learn, share and contribute reaffirms the importance of these initiatives. It’s through them that we not only get a better understanding of the current state of AI, but nurture the next generation of thinkers and innovators. This experience underscored the value of being part of a community focused on learning and growth as much as technological progress.
In this blog post, I will provide an overview of my experience at Hack AI. First, I will describe the format of the event, sharing the prompts and tracks that teams could choose from. Next, I will describe my experience as a mentor. Then, I will highlight the top three winning projects from each track. Finally, I will conclude with some closing thoughts on the event and key takeaways.
Hackathon Details: Prompts and Tracks
The participants had the option of choosing between two tracks. Each track presented a unique challenge, encouraging participants to apply their AI knowledge and skills creatively.
Main Track:
The Main Track challenged participants to create an application integrating a model, which could be an API, like OpenAI’s API, or an model such as Yolo v5 or Teachable Machine. The emphasis was on innovation and practical implementation, allowing contestants to showcase their ability to incorporate advanced AI into functional applications.
AI Club Track:
In the AI Club Track, participants were tasked with developing a machine learning model to predict housing prices from a provided dataset using given data points such as the area of the house and the number of bedrooms. This track aimed to blend data science with real-world applications, highlighting the analytical and predictive capabilities of AI in the housing market.
Deliverables:
The deliverables were clearly outlined for participants to follow. Each team was required to submit their code, either through a GitHub link or a Google Colab link, with a README.md file including the team name and member names. Additionally, a presentation was to be prepared for judging, with in-person presentations for students at OSU and Discord presentations for those outside the university. The presentations were constrained to three minutes, followed by a three-minute Q&A session, making concise communication and clarity paramount.
Reflections on Mentorship at Hack AI
The day started with me delivering a keynote that allowed me to introduce 99P Labs to participants and get them excited for the event. I emphasized what an interesting time it is for them to be pursuing AI, as progress and innovation in the field are happening faster than ever before. New models and tools are released almost weekly. After my talk, I spent most of the day walking around, answering Python questions, explaining concepts like regression, and listening to project idea pitches. Meeting eager students was incredibly fun and rewarding. My mentoring continued into the evening, as I helped teams on Discord with Seaborn visualizations and debugging Haystack code. Though challenging at times, guiding teams was a valuable learning experience for me. I left feeling inspired by the talent and motivation of the next generation of AI innovators. Now let’s look at the outstanding projects that emerged from the event’s two tracks.
The Best of the Best: Top Projects from Hack AI
The creativity on display at Hack AI was truly impressive. The AI Club received over 30 interesting and technically-strong submissions. With so many great projects, selecting only six to recognize was difficult. Though I wish we could have highlighted more teams’ contributions, it simply wasn’t feasible. The standout projects covered next showed exceptional problem-solving, teamwork, and communication. While all participants demonstrated admirable effort and skills, these select few rose above the rest. It was rewarding to see the hackers push boundaries and produce inspiring AI implementations. The winners showed mastery of both the analytical and creative application of artificial intelligence. While they stood apart, the entire group’s passion left me optimistic about the future of AI development. Now let’s look at the top three projects in each track that made a lasting impact.
The Main Track
Within the Main Track, the focus was on developing applications that integrate with models, showcasing innovation and practical application of AI. Participants displayed commendable proficiency in this domain, and the top three projects were exemplary in their innovative approach and execution.
Amove — 1st Place
Securing first place was the project named ‘ZoomGage’ by team Amove, consisting of members Nikhil, Amogh, Kathir, and Aneesh. ZoomGage is an application that uses computer vision to analyze student engagement during remote learning sessions. Its objective is to provide a tool that can automatically give class engagement feedback to teachers without singling out individual students. The project stood out for its potential to enhance the remote education experience by giving educators a means to gauge the effectiveness of their teaching methods in real time. The prize for the first place was $400, to be divided among the team members.
SVAK — 2nd Place
In second place was team SVAK, with members Akshay, Saket, Vihaan, and Karun, who developed ‘BookWork’. This educational tool is designed to aid younger students in enhancing their reading and comprehension skills, with teacher oversight. It demonstrates a practical application of AI in education, specifically targeting the development of foundational literacy skills. The team was awarded a prize of $300 to be shared among its members.
Wild Spirits — 3rd Place
Third place was awarded to team Wild Spirits, comprised of Daniil and Alex, for their ‘SeeFood’ mobile application. SeeFood is an innovative tool that aims to revolutionize the management of dietary needs and preferences. By leveraging cutting-edge technology, it offers users a personalized way to address their dietary choices. The project was commended for its user-centric design and potential impact on individual health management. The team received $200 as their prize.
These projects highlight the breadth of applications for AI, from enhancing educational experiences to personal health management. The winners of the Main Track at Hack AI not only demonstrated technical skill but also the ability to apply AI in ways that can significantly benefit society.
The AI Club Track
In the AI Club Track, participants were challenged to develop a machine learning model to predict housing prices using a set of given data points. This task combined data science with real-world applications to demonstrate the predictive power of AI in the real estate market. It required participants to engage in comprehensive data analysis, apply feature engineering, and optimize machine learning algorithms to achieve the most accurate predictions.
1st Place — Team Alvin
Team Alvin achieved first place with a project focused on housing price prediction using Mean Squared Error (MSE) for their regression model. Unfortunately, the specifics of their methodology and approach are not clearly outlined in the pitch deck they provided. Despite this, their effective use of MSE as a metric suggests a focus on minimizing prediction errors, which likely contributed to their top performance in the competition.
2nd Place — Team Dataholics
Team Dataholics, led by Gowrav Mannem, claimed second place with their project on house price prediction. The dataset provided contained 12 features, and the team undertook an extensive exploratory data analysis (EDA) process. They cleaned the data by removing outliers and applied feature engineering techniques like one-hot encoding for furnishing status and label encoding for other categorical variables. They also addressed multicollinearity by removing variables with high Variance Inflation Factor (VIF) scores. The team used the LazyPredict library to identify the best base model, which turned out to be Random Forest Regressor, and then fine-tuned it using RandomSearchCV and GridSearchCV to improve accuracy and reduce the root mean square error (RMSE).
3rd Place — Team Debug Thugs
The third place was secured by Team Debug Thugs. The specifics of their approach to the housing price prediction challenge are not provided in the documents, but their model would have been evaluated on how well it predicted the prices based on the features of the houses, its generalizability, and the innovation in their methodology.
These projects exemplify the application of AI and machine learning to real-world problems, providing valuable learning experiences for participants and practical solutions that could potentially be translated into real-world scenarios. The AI Club Track showcased the analytical skills and technical expertise required to make precise predictions in a dynamic market such as real estate.
The Projects
As we conclude this showcase of the six winning teams from Hack AI, it’s clear that the event was a testament to the innovation and skill burgeoning in the field of AI. The Main Track winners demonstrated a keen ability to integrate AI into practical applications, addressing challenges in education and personal health management. Meanwhile, the AI Club Track highlighted the analytical depth and predictive prowess of AI in the real estate domain.
The diversity in project themes and the technical expertise displayed by all the winners is indicative of the growing versatility and influence of artificial intelligence. These projects are not just academic exercises; they represent real-world potential and the promise of AI to make significant contributions to various sectors.
It’s uplifting and inspiring to witness such a display of talent and dedication. The participants’ collective effort underscores a bright future for AI, characterized by an eagerness to tackle complex problems and push the boundaries of technology.
Congratulations to all the teams that participated, and a special commendation to the winners for setting a high bar with their exceptional work. Their achievements serve as a benchmark for future hackathons. Hack AI was undoubtedly a melting pot of ideas that will fuel innovation long after the event’s conclusion.
Conclusion
Hack AI demonstrated the vibrancy and potential of The Ohio State University’s AI community. Through dedication and teamwork, participants showcased innovative applications of artificial intelligence across sectors like education, healthcare, and real estate. Their projects exemplified how AI can be harnessed to enhance human experiences and address real-world needs.
Equally important were the connections and conversations fostered at the event. The collaborative atmosphere nurtured learning and growth, enabling participants to share knowledge and push one another to excel. Mentorship played a pivotal role as well, allowing experienced professionals like myself to guide burgeoning talent.
Ultimately, Hack AI embodied the spirit of community building around AI advancement and education. The hunger for hands-on learning and the development of practical solutions revealed a generation eager to drive progress responsibly. Events like this catalyze the future of AI by bringing together diverse perspectives, skill sets, and aspirations under a common purpose.
The passion on display inspires optimism about AI’s trajectory. However, realizing the full potential of these innovations requires ongoing dialogue and collaboration. If you found this glimpse into the hackathon rewarding, I invite you to stay connected with the community. Consider subscribing to the 99P Labs blog, connecting on LinkedIn, or reaching out to explore partnership opportunities.
Through open exchange of ideas, we can direct incredible capabilities toward the experiences that matter most. Technological hype frequently loses sight of human priorities. But working together, we can ensure AI’s awesome potential improves lives. There are always more perspectives to explore and connections to make. 99P Labs looks forward to continuing the conversation with you.
