Inspiration
DreamPath was born from a powerful idea: the gap that exists between having an ambitious dream and knowing exactly what to do on a Monday morning to achieve it. Inspired by the Gabby’s community (led by Gabby Beckford of Packslight), the application seeks to channel that "delusional confidence" into concrete action plans. We wanted to create a tool that wouldn't just motivate you with affirmations, but would use Artificial Intelligence to act as a personal coach that breaks down the impossible into 15-minute steps.
What it does
DreamPath is an "offline-first" personal growth companion that:
- Generates AI Roadmaps: Transforms any dream (from "Learning to Surf" to "Launching a Startup") into a structured 4-week plan with micro-actionable daily tasks.
- Gamifies Progress: Uses the "Jar of Dreams" concept to visualize daily wins, allowing users to level up and earn "Dream Achieved" trophies.
- Boosts Mindset: Offers daily affirmations and personalized "Pep Talks" based on the user's context (Travel, Career, Finance).
- Social Connectivity: Allows users to share their daily wins, weekly roadmaps, and final dream completions on social media to inspire their community.
- Smart Synchronization: Works fully offline and syncs data with Firebase transparently whenever a connection is available.
- Premium Model with RevenueCat: Provides a limited free experience and a premium subscription to unlock unlimited dreams and exclusive coaching content.
How we built it
We built DreamPath using a modern and scalable stack:
- Frontend: Flutter for a fluid cross-platform experience and animations enriched with
flutter_animate. - State Management: Riverpod for robust, reactive, and testable state management.
- Backend & Auth: Firebase (Auth, Firestore, Cloud Messaging for notifications, and Storage for profiles).
- AI Engine: OpenRouter (GPT-OSS-120B) integration for generating plans in structured JSON format.
- Monetization: RevenueCat to manage in-app purchases and the paywall efficiently.
- Architecture: We implemented an Offline-First pattern using a persistent operation queue to ensure no progress is lost without a connection.
Challenges we ran into
One of the biggest challenges was Prompt Engineering. Ensuring the AI consistently returned a valid JSON with a stable weeks/days structure required many iterations. Also, the offline-first synchronization logic was complex: managing conflicts between local data (SharedPreferences) and the cloud database (Firestore) so the user never experiences lag or data loss when switching devices or losing signal.
Accomplishments that we're proud of
We are especially proud of:
- The "Jar of Dreams" visualization: Creating an interface that feels alive and visually rewards every small victory.
- AI Integration: The app's ability to inject personalized "Gabby Tips" into AI-generated plans, making the experience feel human and aligned with the community's voice.
- Paywall Fluidity: A RevenueCat integration that allows for a smooth, friction-less transition between free and premium users.
What we learned
This project was a great school for:
- Mobile App Architecture: We deepened our knowledge of Riverpod Notifiers and professional data persistence.
- Generative AI: We learned to treat LLMs as structured data engines, not just text generators.
- User Experience (UX): The importance of micro-celebrations (confetti, sounds, trophies) in maintaining user retention in productivity apps.
What's next for DreamPath
The future of DreamPath includes:
- Social Dreams: Allowing users to share their roadmaps and progress in a Gabby's Fam community feed to increase accountability.
- AI Mentor Chat: An interactive chat where users can ask specific questions about their current roadmap.
- Wearable Integration: Quick notifications and check-ins from Apple Watch and Android Wear to complete "15-minute" tasks without opening the phone.
Log in or sign up for Devpost to join the conversation.