NINA | Devpost

Home Screen

Inspiration

In today's digital age, we have witnessed the struggles faced by our grandparents and others when navigating computers and technology. Traditional methods of interacting with computers through command-line interfaces (CLIs) and graphical user interfaces (GUIs) can be unintuitive and challenging for those who are not tech-savvy. As society shifts towards natural language inputs for information retrieval, we recognized the opportunity to extend this approach to file management. Inspired by the desire to make technology more accessible, we developed NINA (Natural Interaction and Navigation Assistant). Our goal is to empower users, regardless of their technical background, to effortlessly organize and work with their files using natural language commands, fostering a more inclusive digital environment where everyone can thrive.

What it does

NINA is an AI-powered file assistant that revolutionizes file organization. With simple prompts, NINA intelligently analyzes, categorizes, and arranges your files, saving you time and effort. It learns your unique content patterns and folder structures to provide a personalized file management experience.

How we built it

We built NINA using advanced large language models and natural language processing techniques. We collaborated to create a robust and intuitive system that understands user intent and adapts to individual preferences. NINA's backend is powered by a combination of the most powerful AI language models and a sophisticated indexing algorithm, ensuring accurate and efficient file organization. We build our entire application in Swift, creating a native MacOS application.

Challenges we ran into

One of the main challenges we faced was dealing with folder access permissions and a vast diversity of file types, and organizational structures across different users. We had to develop sophisticated algorithms that could handle this variability and provide consistent results. Additionally, ensuring the security and privacy of user data was a top priority, requiring us to implement strong encryption and access controls. The tech stack was also a challenge for us as although we decided to use React Native initially, we quickly realized its limitations for handling local files and had to pivot to building a native app with Swift, which none of us had prior experience with.

Accomplishments that we're proud of

We are proud of creating an AI-powered file assistant that genuinely simplifies file management for users. NINA's ability to understand natural language prompts and quickly organize files is a significant accomplishment. We are also proud of the system's adaptability and learning capabilities, which allow it to improve over time based on user interactions.

What we learned

We were all new to native MacOS development with Swift so it was a great experience for us to learn this framework. We also got extensive experience in working with LLMs and Shell scripting in order to be able to understand and execute user queries.

What's next for NINA

We think the future of interacting with computers is natural language and the large goal here is an LLM powered operating system. NINA is a proof of concept and a fundamental building block of this dream. We hope to develop NINA further and bring it to the computers of those who want a new way of interacting with their files!