BayEye Briefings

AI meets SF civic pride with BayEye Briefings—hyper-personalized newsletters that turn local insights into daily discoveries.

Comment

Alt video link, containing audio

High quality example copies of BayEye Briefings:

BayEye on SF Community

BayEye on FiDi

BayEye on SF Supervisors

BayEye on SF Transportation

BayEye on Clean Safe SF

Inspiration

As someone born & raised in San Francisco, I'm driven by a commitment to uphold the authenticity of our city's narrative, and provide accessible data-driven insights to citizens on topics they are interested in. BayEye Briefings was born from the vision of using AI to sift through complex datasets and deliver nuanced, personalized stories that make data relatable and impactful for individuals across our city.

What it does

BayEye Briefings is an innovative application that uses the Scale AI Donovan platform to generate highly personalized newsletters. It analyzes current events, public data sets from SF Gov, and user-provided interests to deliver tailored content that informs, engages, and connects San Francisco residents with their city. Users receive updates on everything from local parks and recreation to food safety inspections, all contextualized to their lifestyles and preferences, thanks to the intelligent parsing of AI.

How we built it

Our stack is built on Next.js and React, leveraging open data sets provided by SF Gov, fed into GPT3.5-turbo through Donovan to generate the content for the newsletters.

Challenges we ran into

One of the most significant challenges was ensuring the accuracy and relevacy of the information being parsed and presented. We had to refine our LLM queries to filter out noise and focus on data that would be most beneficial to the user. Another hurdle was designing a system that could provide personalized content without compromising user privacy or data security.

Accomplishments that we're proud of

I'm proud of creating a platform that not only informs but also connects citizens with the heartbeat of their city in a way that is engaging and personal. This project bridges the gap between raw data and actionable knowledge, facilitating a well-informed community.

What we learned

Through this project, we've gained a deeper understanding of the technical challenges involved in processing large data sets with AI and the design considerations needed to create a user-friendly interface that can deliver complex information simply. We've also learned about the potential and limitations of LLMs in curating content that's both accurate and personal.

What's next for BayEye Briefings

Looking ahead, we plan to expand BayEye Briefings to cover more topics and integrate with more data sources for even more personalized content. We will explore ways to enhance the AI's understanding of user feedback to refine content accuracy and relevancy further. Long-term, we aim to replicate our model for other cities, promoting informed civic engagement on a global scale.

Built With

Share this project:

Updates