SlugLine

Inspiration

As engineering students, we encounter countless clubs, events, and opportunities, but the announcements are scattered across email, Slack, Discord, and Instagram. Keeping up means constantly switching platforms and searching through unrelated messages. We wanted one place that answers three simple questions: Who is hosting it? Where is it happening? When do I need to act?

What it does

SlugLine brings campus opportunities into a personal dashboard, with a calendar for dates and deadlines and an interactive map for locations. Our prototype includes the dashboard, calendar, and mapping interface, with initial Gmail connectivity.

Our goal is to let students connect their own accounts and use AI to turn relevant messages into organized opportunities. Each result would include the organizer, dates, location, signup link, and supporting source information.

How we built it

We started with the dashboard, calendar, and map, then worked on connecting individual users’ Gmail accounts through OAuth.

We also designed a multi-agent AI workflow: separate agents would process Gmail, Slack, and class schedules, while verification and matching agents would check source evidence, identify duplicates, and flag schedule conflicts. A coordinator would manage these tasks and bring their results together.

Challenges we ran into

Connecting personal accounts was more complicated than creating a login button. We encountered OAuth testing restrictions, permission requirements, and differences in what each platform allows applications to access.

Another challenge was defining trustworthy AI behavior. Event dates and application deadlines must remain distinct, missing details should stay unknown, and duplicate announcements should not overwhelm the dashboard. These requirements shaped our processing workflow.

Accomplishments that we're proud of

We turned a problem from our own student experience into a working visual foundation with a calendar and interactive map. We also established initial Gmail connectivity and developed a concrete plan for processing information from multiple sources.

Most importantly, we kept the project focused on each student’s own information and interests.

What we learned

We learned that authentication, data access, and AI processing are separate parts of an application. Connecting an account does not automatically fetch its messages or turn them into useful results.

We also learned that a useful AI workflow needs clear responsibilities, source evidence, privacy boundaries, and understandable error messages.

What's next for SlugLine

Our next milestone is collaborating with SlugPath - a UCSC events AI chatbot being developed by Tech4Good

Also, a complete Gmail workflow: scan authorized emails, extract opportunities, verify their details against the messages, and let users review and save the results.

After that, we plan to add Slack/Discord scanning, class schedule imports, duplicate detection across sources, and personalized relevance ranking. We also want to make account connection smoother and prepare the application for broader access beyond test users.

Built With

• agentic
• api
• css
• html
• javascript
• node.js
• react