What it does
Ideally:
The user enters their state and county, as well as topics of interest in their community. Gemini then returns a set of goals and tactics in the form of a matching game. The user pairs them as they see fit and when finished their pairings are assessed by Gemini to see if they pair well (ie. if you want to raise minimum wage, striking may be more of an effective strategy than a fundraiser). Then the user is given the option to see the optimal pairings, and learn more about the goals and methods. Every step of the way Gemini is providing sources so the user knows they can trust the information, and continue research on their own if they are so inclined.
Inspiration
Many of my close friends and family have become activists and I often find myself wondering if we are doing enough. After learning what makes effective activism I wanted to not only share but help personalize these strategies. An app named 5 calls was a large inspiration for this project.
Where I'm at right now:
- interfaces with Gemini API to provide context appropriate items for users
- gamifies the process of learning about effective ways to be an activist
How I built it
- Flask server to interface with GPT (and store user info in the future)
- React front end - spotlighted library: dnd-kit
Challenges I ran into
- learning how drag and drop worked - not intuitive
- Gemini:
- API key - use AI studio or Open router???
- OUTPUT FORMATTING
Accomplishments that I'm proud of
What I learned
- How to create a more realistic time table - continuous reality checks & reprioritizing
- Became fairly well versed in many features in Gemini
- How to read documentation faster
What's next for Assemble
- Perfecting the main feature: Interests --> Mix and match --> Learning
- Extend off that: Option to build a short + long term action plan?
- Community feature: Maybe something along the lines of "there are 19 other people in your area who also want better access for bikes, want to connect?"
Log in or sign up for Devpost to join the conversation.