Inspiration
We started Haply from a simple observation: personal coaching is powerful, but it is often expensive, inaccessible, or difficult to maintain consistently. Many people want guidance for building habits, improving routines, or staying accountable yet most tools either overwhelm users with complexity or fail to adapt to individual needs.
We wanted coaching to feel natural, conversational, and always available. Instead of rigid programs or static trackers, we envisioned AI coaches that guide users through everyday life in a practical and supportive way. The goal was not just to simulate conversation, but to turn guidance into real-world action habits, streaks, routines, and measurable progress.
Haply is our attempt to bridge intelligent assistance with daily behavior change, making personal coaching accessible without friction.
What it does
Haply is an AI-powered personal coaching platform that helps users build habits, structure their day, and track progress through conversational guidance.
Users can explore specialized AI coaches, chat in real time, and receive personalized recommendations that extend beyond conversation. Coaches can suggest habits or routines that users approve and track inside the app. A built-in dashboard supports weekly scheduling, habit streaks, XP progression, and daily motivation transforming coaching into a continuous system rather than a one-time interaction.
A key feature is coach creation: users can design their own AI coach by defining tone, behavior, and purpose. This enables a personalized coaching ecosystem where guidance feels relevant and human-centered.
How we built it
Haply is built as a cross-platform mobile application using Flutter, delivering a consistent experience across iOS and Android. We structured the app with a modular architecture that separates coaching logic, habit systems, and interaction layers to support scalability and rapid iteration.
At the core of Haply is the Gemini API, which powers the intelligence behind every coaching interaction. Rather than using Gemini as a simple chat generator, we designed it as the decision engine that drives personalized coaching behavior. Gemini interprets user intent, coach configuration, and conversation context to produce structured, actionable outputs — such as habit suggestions, planning systems, and user-defined app logic.
We implemented a structured prompting pipeline that allows Gemini to operate within defined coaching roles. This ensures responses remain consistent, goal-oriented, and aligned with each coach’s purpose. Gemini also supports the Apps system, where users collaborate with their coach to build practical mini-tools like planners or trackers that extend beyond conversation.
State management and routing are handled through GetX, enabling reactive UI flows and efficient dependency management. Backend communication runs through REST APIs to coordinate AI sessions, persistence, and real-time updates.
Challenges we ran into
One of the biggest challenges was translating conversational AI into actionable behavior systems. Chat alone is not enough we needed a framework where AI suggestions could become structured habits without overwhelming users.
Balancing automation with user control required careful design. Users should feel guided, not dictated to. Creating approval flows, streak tracking, and scheduling inside a chat-driven experience required multiple iterations.
Maintaining responsiveness while handling asynchronous AI communication and UI updates was another engineering challenge. We also designed a coach ecosystem flexible enough to support user-created coaches without sacrificing clarity or usability.
Accomplishments that we're proud of
We are proud of turning AI coaching into more than just conversation by building a system where guidance becomes usable tools. Haply connects coaching, habits, streaks, and daily planning into one cohesive experience that users actively interact with.
Our most valuable accomplishment is the Apps system. Instead of limiting users to fixed features, Haply allows them to build their own mini applications with the help of their AI coaches. Users can design tools tailored to their needs such as custom planners, trackers, or routines where structure, behavior, and purpose are shaped by the user. This transforms coaching into something practical and personal rather than passive.
The coach creation system further extends personalization, allowing users to define how guidance is delivered. Combined with user-built apps, Haply becomes a flexible ecosystem where coaching adapts to individual workflows.
Most importantly, Haply delivers immediate value: users talk with their coach, build tools around their goals, and turn daily actions into measurable progress.
What we learned
Building Haply taught us that behavior change requires more than intelligence it needs structure, simplicity, and emotional engagement.
We learned that conversational interfaces must be paired with clear action systems. Users want guidance, but they also want autonomy and visibility into progress.
Technically, we strengthened our understanding of real-time mobile architecture, scalable state management, and designing systems where AI integrates seamlessly with user experience.
Iteration proved critical small UX decisions significantly impact motivation and consistency.
What's next for Haply - AI Personal Coaching App
Our next focus is making coaching interactions feel more natural and immersive. We are developing avatar-supported video coaching, where users can interact with their AI coach through a visual interface that simulates face-to-face guidance. The goal is to move beyond text chat and create a more engaging coaching experience while maintaining responsiveness and accessibility.
We are also expanding the Apps system so users can build richer, coach-assisted tools tailored to their workflows such as planning systems, progress trackers, and goal frameworks directly inside the platform. This will allow coaching conversations to evolve into practical systems users rely on daily.
In the near future, Haply will transition from a prototype coaching environment into a market-ready personal coaching platform combining conversational AI, user-built tools, and immersive interaction to support everyday growth.
Log in or sign up for Devpost to join the conversation.