Application Spotlight: Luna
Hey everyone, I’m Luna, a current student at the University of Washington and a hackNY 2024 fellow. I had an amazing time in the fellowship this past summer, and I especially loved the community. A few of us even had a mini reunion at Hack The North! I wanted to share my application for hackNY in hopes of helping any aspiring applicants. Even if you don’t feel qualified, you should always apply! I personally didn’t think I’d get in but decided to go for it anyway. The most important thing is to be yourself and let your voice and passions shine through in your application. While my application definitely isn’t perfect, I hope it can provide some guidance, encouragement, and a bit of insight into the process for anyone considering applying. Good luck and if you have any questions feel free to reach out!
Tell us about a coding project or a developer community project you’ve built that you are most proud of:
This project is an AI-powered tool for medical interpretation that aims to improve understanding between healthcare providers and patients. The tool offers several key functionalities, including real-time audio recording, speech-to-text conversion, translation, and simplification, enabling communication between medical professionals and patients who speak different languages. To bring this project to life, we utilized various technologies, such as OpenAI for prompt engineering, Google Text-to-Speech for high-quality audio feedback, and FeatureForm for defining our model's features. The end result was a tool capable of transcribing spoken words and providing real-time translations and simplifications, making it invaluable for healthcare providers. For example, a patient could record a doctor's explanation on the platform and receive not only translations but also simplified medical terminology in their native language.
Given the time constraints of the hackathon, we followed a development approach based on a product requirements document. In this document, we brainstormed different features and ranked them based on priority. By prioritizing these features, we ensured that even with limited time, the core functionality would remain intact and usable. Regular check-ins and constant communication among team members ensured alignment with project goals and facilitated quick resolution of any development challenges that arose. We assigned different roles within the team, with one member handling product management, another focusing on machine learning and implementing FeatureForm, and myself and another team member working on the backend and utilizing the OpenAI API. While my teammate focused more on the frontend, I worked on features such as implementing voice recording uploads and determining the most suitable API for text-to-speech functionality. Our main focus was on building a simple prototype by integrating Flask for the backend and incorporating OpenAI's GPT-4 API to transcribe and simplify medical terminology. Despite having no previous experience with artificial intelligence, I overcame this challenge by extensively studying the OpenAI API documentation and collaborating with my teammates. Throughout the development process, we encountered challenges related to time management, necessitating quick decision-making regarding feature prioritization and implementation. We also faced difficulties in handling rate limitations imposed by external services like Google Text-to-Speech. To address these challenges, we relied on the product requirements document to stay focused on priorities, optimized code efficiency to streamline processes, and explored alternative services for text-to-speech functions. This approach enabled us to effectively overcome challenges and ensure the success of the project within the demanding constraints of the hackathon. All in all, I was very happy with the end result especially since most of us were new to working with artificial intelligence. We went into the project with the mindset of learning and growing, and we reached that goal, with an addition of a hackathon prize which was simply the cherry on top.
Why did you build this project? What about it matters to you?
One aspect of this project that I am particularly proud of is the extensive time my team and I dedicated to brainstorming and ensuring that our project addresses a real- world problem where we can make a meaningful impact. We aimed to tackle a critical healthcare communication challenge rather than creating a simple project for a weekend as our goal was to build something that would have a lasting impact.
I built this project because I can personally relate to the struggles of translating for a parent as a child of immigrants. A few years ago, when I accompanied my mom to the doctor's office, language barriers persisted, hindering my mom from fully grasping her healthcare options. As I juggled the roles of translator and simplifier of medical jargon, I keenly felt the inadequacy of the existing system. This shared experience underscored the struggles familiar to many children of immigrants. Motivated by this personal connection, I initiated this project to ensure that individuals in countries where the native language is unfamiliar, can confidently navigate healthcare decisions. My vision extends to a future where families like mine can engage meaningfully with healthcare professionals, comprehending medical nuances and making informed choices for healthier lives. We made our project so that, moving forward, patients in countries where they don’t speak the native language can feel comfortable in the doctor’s office, enabled to make informed decisions about their health, their bodies, and their lives.
I also built this project so I could gain skills in artificial intelligence, an area that is all over the news right now. Despite reading articles about artificial intelligence and all the new artificial intelligence startups emerging, I knew that the best way for me to understand the technology was by creating a project for hands-on experience. I was excited to learn more about prompt engineering as well, and learn the nuances that go into ensuring that you receive the output you want. For example, the initial translation was very convoluted and we learned to be specific in how long the translation should be and how simple we want it to be. After creating this project, I felt a lot more confident in my skills in reading documentation, understanding OpenAI, and the various technologies that go into building an artificial intelligence tool. This project not only allowed me to hone my technical skills but also demonstrated the transformative potential of AI in healthcare, which is an area I'm deeply passionate about.
If you could re-build this project from scratch, what would you do differently?
If I were to rebuild this project from scratch, I would allocate more time for user interviews and feedback iterations. While we were able to develop a functional tool within the time constraints of the hackathon, there were limited opportunities to gather user input and refine the user experience. By involving potential users earlier in the development process, we could gather valuable insights and make iterative improvements based on their feedback.
Additionally, I would explore the integration of more advanced AI models and technologies, such as natural language understanding and sentiment analysis, to enhance the accuracy and effectiveness of the tool in interpreting and translating medical conversations. This would further improve the tool's ability to provide accurate and meaningful translations, ensuring that healthcare providers and patients can communicate effectively and understand each other's needs.
Another aspect I would focus on is scalability and deployment. In the hackathon, we built a prototype of the tool, but to make it a viable solution in real-world healthcare settings, it would be important to consider the infrastructure required for high-volume usage and ensure robustness and reliability. This could involve implementing distributed systems and optimizing resource allocation to handle multiple concurrent users and large amounts of data. Additionally, I would pay attention to security and privacy concerns, ensuring that patient data is protected and compliant with relevant regulations.
Furthermore, I would invest more time in exploring partnerships and collaborations with healthcare institutions and organizations. By working closely with medical professionals and understanding their specific needs and workflows, we could tailor the tool to better integrate into existing healthcare systems and to ensure that the tool meets their requirements and aligns with industry standards.
Overall, by prioritizing user feedback, incorporating more advanced AI technologies, focusing on scalability and deployment, and collaborating with healthcare institutions, I believe we could further enhance the impact and usability of this AI- powered tool for medical interpretation. My team was able to continue working on this project after the hackathon and implement some of the above ideas, which was exciting as we did user research and talked with hospitals in order to better understand if this tool could actually be implemented and if it addressed existing pain points.
Tell us about societal issues that you'd like to solve, why they matter to you, and what you've done to address them․
A house. A shelter. A simple thing, right? No. In high school, I finally realized the extent of the homelessness problem in Seattle. As the President of the Homeless Education Living Project Club, I was able to consciously make an effort to help people who are homeless by creating care packages, making blankets, and hosting fundraisers. Now, when I drive through the city, I notice people living in tents along the highways and holding up signs asking for money. However, instead of feeling guilty and quickly walking past, I give the person a care package; I keep extras from the club in my car. Homeless people often lack a voice in my community, but through this club, I am able to make their voices heard and help others see that one person is enough to make a change. Through this experience, I’ve learned about the danger of a single story when used to define a person, and that has stuck with me.
This passion for helping others has continued onto university, where I’ve found different ways to help with my newfound technical skills. The first coding project I created for a hackathon was a sustainability and food waste app. I also became an officer of my school’s Girls Who Code club and organized two hackathons for marginalized genders, where I was able to help over 1000 hackers. I almost didn’t pursue computer science until I participated in a free two week summer program for female and non-binary individuals so I want to pass it on by helping empower other women in tech. As a result, I am excited about social impact initiatives that focus on improving access to education for underserved communities. Education plays a crucial role in empowering individuals and breaking the cycle of poverty. I would love to participate in social impact initiatives that spread technology to marginalized communities, and also teach individuals about the ethics of tech.
By delving deeper into issues like homelessness, I’ve also found it interesting to dissect what the root problem is. You can make a million care packages to distribute to the homeless, but that will not reduce the issue of people experiencing homelessness as more affordable housing is needed for a more long-term solution. At hackNY, I hope to collaborate with like-minded individuals to create technology- driven solutions that address the root causes of societal issues and ensure the social impact of our projects lasts.
Why apply to hackNY specifically?
The summer before my freshman year of university, I stumbled upon a post about the first all female hacker house in San Francisco. It looked intriguing and after some convincing of my parents, I found myself on a flight to California to live with fourteen other strangers—the first of many spontaneous decisions in the years to come. That summer was transformative. I went into it with no expectations and emerged a more confident knowledgeable individual with a newfound fascination with startups. It was also the first time I met people working all day not out of obligation but driven by a genuine love for what they were creating. The mindset of continuous learning and growth permeated the space, creating an environment where work felt like an exploration of passion. Being in that environment was very motivating and it was also where I learned the importance of building a community, a lesson I carry with me and eagerly anticipate recreating with hackNY.
My commitment to community building has only deepened in university. I am an organizer of my university’s first student-run startup incubator program and work part-time for a venture fund, where my main responsibilities include hosting meetups and bringing people into the community. Now, as I look to connect with hackNY, I am drawn to the prospect of contributing to a community that not only shares my passion for startups and technology, but also for social impact. In a world where AI is advancing at an unprecedented pace, hackNY emphasis on socially conscious programming is more important than ever. I’m especially excited to be in New York as well, a city overflowing with diversity and innovation which has a magical way of facilitating meaningful encounters. Whether it's striking up conversations with other fellows in the late hours or even randomly meeting inspiring individuals on the bustling streets, I see NYC as a canvas for creating lifelong connections. I’m looking forward to deep talks with like-minded individuals, exploring the intricate intersections of technology, ethics, and society. All in all, I'm most excited about the possibility of discovering not just collaborators but friends to accompany me in the journey of life in the years to come (perhaps even finding my future co-founder!).
