Momentum

Momentum AI Co-Pilot: Achieve your goals, one smart step at a time.

Comment

Inspiration In a world filled with productivity tools, we noticed a fundamental gap. We have apps to track our tasks and calendars to schedule our time, but when faced with a truly ambitious, life-changing goal—like learning a new skill, launching a business, or running a marathon—we are often left overwhelmed. The path from aspiration to achievement is unclear, and the initial excitement quickly fades into confusion and burnout. The inspiration for Momentum came from a simple question: What if your planner was not just a passive list, but an intelligent, cognitive partner that could chart the course for you? We wanted to build an app that serves as an AI co-pilot for your goals, providing clarity, adapting to life's unpredictability, and making a daunting journey feel manageable and exciting.

What it does Momentum is an AI-powered life co-pilot that transforms your most ambitious goals into a clear, daily, actionable plan.

At its core, Momentum does the following:

Intelligently Differentiates Tasks: It uses an AI "Smart Gateway" to understand if your input is a simple to-do (like "Send an email") or a complex goal ("Become a Data Analyst"). For simple tasks, it gets out of your way.

Asks the Right Questions: For complex goals, it initiates a dynamic, AI-driven conversation to understand your current skill level, your motivations, and the resources you already have.

Generates Comprehensive Roadmaps: It takes this information and calls on Google's Gemini Flash model to generate a complete, day-by-day plan for the entire duration of your goal. This isn't just a task list; it includes meaningful daily activities, recommendations for books or videos to fill identified "knowledge gaps," and motivational insights.

Provides Daily Clarity: While the full plan is always accessible, the app's main screen is a clean, minimalist view of only today's tasks, preventing overwhelm.

Adapts to Reality: Through its "Resilience Engine," Momentum automatically detects when tasks are missed and offers simple, non-judgmental options to either integrate them into the upcoming schedule or extend the final deadline, ensuring you never feel derailed.

How we built it Momentum is built on a modern, scalable, and AI-native tech stack, chosen for performance and rapid development.

Backend: We used FastAPI (Python) for its incredible speed and ease of use. As the native language of AI, Python allowed for seamless integration with our core AI model.

AI Model: The brain of our application is Google's gemini-1.5-flash-latest model. We engineered a sophisticated, multi-step prompting system: a "Triage" prompt classifies the user's input, a "Dynamic Questionnaire" prompt generates context-aware follow-up questions, and a final "Comprehensive Plan" prompt builds the detailed, long-term roadmap.

Database and Authentication: We chose Supabase as our backend-as-a-service. It provided us with a robust PostgreSQL database to store user goals and plans (using the efficient JSONB data type), along with a secure and easy-to-implement user authentication system right out of the box.

Frontend: The mobile application was built with Flutter (Dart), allowing us to create a single, beautiful, and high-performance codebase for both iOS and Android. This ensured a consistent and clean user experience, which is a core tenet of our project.

Challenges we ran into Building a truly intelligent system presented several unique challenges:

Sophisticated Prompt Engineering: Crafting the AI prompts was an iterative process. It was a challenge to design a prompt that could reliably generate a complete, multi-day plan in a perfect JSON format every time, complete with tasks, resource recommendations, and knowledge gap analysis. We had to build robust error handling and parsing on our backend to manage this.

Maintaining Simplicity (The "Don't Overwhelm" Principle): The biggest design challenge was housing an incredibly powerful planning engine within a user interface that feels simple, clean, and calming. We adopted a "progressive disclosure" philosophy, ensuring that the user only ever sees what they need for the present moment (today's tasks), while keeping the full, complex plan accessible but tucked away.

The Logic of Rescheduling: Designing the "Resilience Engine" was complex. The logic to intelligently "sprinkle" missed tasks into a user's future schedule without creating an overwhelming new schedule required careful planning on both the backend (to update the plan) and the frontend (to clearly communicate the change to the user).

Accomplishments that we're proud of We are incredibly proud of creating an application that feels more like a partner than a tool. Specifically:

We successfully built an AI system that goes beyond simple task generation to create truly comprehensive, long-term roadmaps tailored to the user.

The "Smart Gateway" that differentiates between simple to-dos and complex goals works seamlessly, providing a frictionless user experience.

The "Resilience Engine" is a feature we believe truly sets Momentum apart. By handling setbacks with empathy and intelligence, it directly tackles the main reason people abandon their goals.

We achieved our primary design goal: creating an app that is powerful and deeply useful without ever feeling overwhelming to the user.

What we learned This project was a tremendous learning experience, both technically and philosophically.

The Power of the Prompt: We learned that the quality of the AI's output is almost entirely dependent on the quality and structure of the prompt. A well-engineered prompt is as important as well-written code.

User Experience is Everything: An intelligent backend is useless if the frontend is cluttered. We learned the importance of ruthless prioritization in design, focusing on the user's immediate needs to build a truly helpful product.

Empathy in Software: We learned that for a productivity app, how you handle "failure" (like a missed task) is more important than how you handle success. Building an empathetic, non-judgmental system is key to long-term user engagement and real-world success.

What's next for Momentum This first version is just the beginning, proving the core vision. Our roadmap is focused on making Momentum an even more indispensable co-pilot for life:

The "Skill Graph": We plan to build a system that visually tracks the specific skills a user is learning, creating a beautiful "tree" of knowledge that grows as they do.

Deeper Integrations: We want to connect Momentum with calendars (to schedule tasks in real-time), note-taking apps, and third-party learning platforms.

Community & Accountability: Introducing features for users to share their goal progress (not their tasks) with friends for accountability, and creating community hubs where users with similar goals can share resources and advice.

Proactive Mentorship: Enhancing "Aura" to be more proactive, suggesting adjustments to the plan based on the user's progress and even warning of potential burnout based on their activity patterns.

Built With

  • expo-managed-workflow
  • expo-router
  • expo-sdk
  • google-gemini-ai
  • lucide-react-native
  • react-native
  • react-native-reanimated
  • revenuecat
  • stylesheet
  • supabase
  • typescript
Share this project:

Updates