Inspiration
I tried thinking about what would be useful and thought about experiences I had on campus. For example I don't know the schedule of every single dining location on campus or sometimes I did not know where I could get a specific thing I wanted to eat. I would have to physically search up on the Stony Brook dining page the hours and details of a certain dining location.
What it does
SBU Dining agent is an Ai assistant that could help anybody that goes to Stony Brook with dining options, dining hours and if a certain location is closed based on any extreme weather.
How we built it
I tried following a lot of the Ai workshop examples on my own. Similar to the workshops I used a Groq API, openai/gpt-oss-20b, and an LLM. Additionally I gave my ai agent a json file for it to use as local information.
Challenges we ran into
The main challenge I faced was an API error where the search tool returned a python list instead of a string. The web page search was the most challenging because of that.
Accomplishments that we're proud of
I am proud for making a project that works, creating my first Ai project.
What we learned
Throughout the workshops and through the project I learned a lot and implemented a lot of concepts like RAG.
What's next for SBU DINING AGENT
For this SBU Dining Agent, it could be expanded to be more specific for nutritional needs or like helping find the location in the moment if the person can't find it.
Log in or sign up for Devpost to join the conversation.