Fresh Air.

Fresh Air Opener
Select Your Preferences to Tailor Your Mission
Interact With The AI and Generate Your Mission
Save Your Progress and Feedback For The AI's Memory

Inspiration

I wanted to explore how AI could be used for something beyond keeping people on their screens. Sometimes, when we're stressed, overwhelmed, or stuck in our own heads, stepping outside and doing something small can help us reset. That inspired Fresh Air and its tagline: Get out your feelings and get outside! I wanted to build something that uses technology to encourage people to reconnect with themselves and the world around them.

What it does

Fresh Air generates personalized outdoor missions based on users' preferences and goals. Users can explore alternative activities, translate mission instructions, complete missions, and track their progress in a journal. The goal is to make it easier to find simple, meaningful things to do offline while encouraging personal well-being and environmental awareness.

How we built it

I built Fresh Air using Python and Streamlit, with Ollama powering AI-generated missions. I integrated MongoDB Atlas for persistent journal data, SerpApi for information retrieval, and ElevenLabs for voice narration. Each technology serves a different purpose, helping connect mission generation with the features users interact with throughout the application.

Challenges we ran into

Getting multiple technologies to work together was one of the biggest challenges. I had to troubleshoot integrations, resolve configuration and model issues, and test the application's features to make sure they worked as expected. I also had to think about how to keep the experience simple without losing the functionality that makes Fresh Air useful.

Accomplishments that we're proud of

I'm proud to have turned an idea into a working application that brings together AI, personalization, and real-world activities. Instead of stopping at generating a response, Fresh Air encourages users to actually do something with it. I'm also proud of the different features working together to create an experience centered on spending less time online and more time engaging with the world.

What we learned

I learned that building an AI application involves much more than choosing a model and writing prompts. It requires connecting different services, testing features, troubleshooting problems, and thinking about how people will actually use the finished product. Most importantly, I learned to focus on what the technology helps someone accomplish rather than adding features just because they are possible.

What's next for Fresh Air

I want to keep improving how Fresh Air personalizes missions, learns from user feedback, and tracks progress over time. I'd also like to make the experience more accessible and provide better alternatives when a suggested activity isn't practical for someone's location, resources, or circumstances. Ultimately, I want Fresh Air to help people build small habits that get them outside, give them a break from their screens, and encourage them to care for their surroundings.