Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments thaInspiration

We've all been there — staring at a to-do list that grows faster than we can check things off, setting New Year's resolutions that quietly die by mid-February, and feeling like we're always busy but never actually productive. The tipping point came when one of our teammates, a chronic over-planner, realized she had 14 different apps for tracking habits, tasks, mood, sleep, and goals — yet still felt completely lost. We asked ourselves: what if there was one intelligent companion that understood your patterns and actually helped you become the best version of yourself?

That's how "I'm number 01" was born — not just another habit tracker, but a personalized AI growth engine that adapts to you.

What it does I'm number 01 is an AI-powered personal growth platform that combines habit tracking, goal management, and intelligent coaching into a single, beautiful experience. Here's what it does:

Smart Habit Tracking: Log your daily habits (reading, exercising, meditating, you name it) and the app automatically identifies patterns in your behavior — like that you're 73% more likely to skip gym on rainy Tuesdays. AI Coach "Zero": A conversational AI assistant that checks in with you daily, offers personalized encouragement, adjusts your goals based on your streaks and slip-ups, and even sends "tough love" nudges when you're slacking. Mood & Energy Correlation: By integrating optional wearable data and quick mood check-ins, the platform surfaces insights like "Your productivity peaks between 9–11 AM on days you sleep before 11 PM." Growth Timeline: A visual journey of your progress — streaks, milestones, setbacks, and comebacks — rendered as a beautiful, scrollable timeline. Social Accountability Circles: Join small groups of 3–5 people with similar goals. The AI matches you based on commitment patterns and time zones, fostering genuine accountability without the noise of larger communities. How we built it Our tech stack is a mix of battle-tested tools and cutting-edge AI:

Frontend: React + TypeScript with Tailwind CSS for a clean, responsive UI. We used Framer Motion for fluid micro-interactions that make habit-tracking feel satisfying. Backend: Node.js with Express, backed by PostgreSQL for structured data (users, habits, goals) and Redis for caching and real-time notifications. AI Engine: We fine-tuned a GPT-based model on behavioral psychology frameworks (CBT, habit loops from Atomic Habits) to power the "Zero" coach. The model generates context-aware check-ins, interprets mood patterns, and suggests micro-adjustments to goals. Data Pipeline: A Python-based analytics service running scheduled jobs to compute streak probabilities, correlation matrices between habits/mood/weather, and personalized insights. Wearable Integration: OAuth-based integration with Apple Health and Google Fit APIs for optional sleep and activity data. Infrastructure: Dockerized services deployed on AWS (ECS + RDS), with GitHub Actions for CI/CD. Challenges we ran into AI Coach Tone Calibration: The hardest part was making "Zero" feel genuinely supportive rather than annoying or robotic. Early versions were either too cheerful (felt fake) or too clinical (felt cold). We ran 200+ A/B tests with beta users and iteratively adjusted the system prompt and temperature settings until we landed on a tone that users described as "a friend who actually gets it."

Correlation vs. Causation: Building the insights engine, we had to be extremely careful not to present spurious correlations as actionable advice. We implemented a statistical significance threshold (p < 0.05) and added disclaimers to every insight. It was a humbling reminder that "data-driven" doesn't automatically mean "correct."

Real-time Streak Calculations at Scale: Computing habit streaks across thousands of users with varying time zones and custom schedules turned out to be surprisingly complex. We ended up building a custom time-window engine in Rust and exposed it as a microservice — a 40x performance improvement over our initial Node.js implementation.

Designing for Consistency Without Guilt: We wanted the app to motivate, not shame. Designing the UI/UX to celebrate comebacks as much as streaks took several redesigns. The final "Comeback Bonus" feature — which gives bonus points for resuming a habit after a break — was directly inspired by user feedback.

Accomplishments that we're proud of Working MVP in 36 hours with a fully functional AI coach, habit tracker, and insights engine — all deployed and accessible online. 94% positive sentiment from our demo day users, with several asking "when can I download this?" The "Comeback Bonus" feature, which we believe is genuinely novel in the habit-tracking space and directly addresses the #1 reason people abandon habit apps: the guilt of breaking a streak. A clean, accessible UI that passed WCAG 2.1 AA compliance checks on the first audit — something we prioritized from day one. Zero's personality — we managed to give an AI coach a distinct, warm, and slightly witty voice that users genuinely looked forward to interacting with. What we learned Behavioral psychology is hard: You can't just throw AI at human motivation and expect magic. Understanding why people form (and break) habits required us to dive deep into research — from BJ Fogg's behavior model to dopamine reward loops. The tech was the easy part; the human part was the real challenge. Tone is everything in AI products: A 1% change in the AI coach's wording could be the difference between a user feeling inspired and a user uninstalling the app. We learned to treat language as a first-class engineering concern. Edge cases in time zones will haunt you: "Did I break my streak?" is a question that sounds simple until you realize your user just flew from Tokyo to Los Angeles and technically gained a day. Ship fast, but ship right: We initially cut corners on input validation and paid for it with a 2-hour debugging session at 3 AM. Lesson learned: even in a hackathon, do the boring stuff properly the first time. What's next for I'm number 01 Native mobile apps (iOS & Android) using React Native — because habit tracking needs to be in your pocket, not just your browser. Proactive AI interventions: Instead of waiting for users to open the app, "Zero" will intelligently send push notifications at your personal "high-willpower moments" based on your historical patterns. Gamification layer: A points system, badges, and seasonal challenges — but designed carefully to enhance intrinsic motivation, not replace it. Open API & integrations: Connect with Notion, Google Calendar, Spotify (for focus music), and more — making "I'm number 01" the central hub for personal growth. Longitudinal research partnership: We're in early talks with a university behavioral science lab to study the app's real-world impact on habit formation over 6+ months. The big vision: Become the operating system for personal growth — the one app that truly understands who you are, who you want to be, and walks with you every step of the way.t we're proud of

What we learned

What's next for I’m number 01

Built With

  • fine-tuning
  • language
  • natural
  • openai-gpt
Share this project:

Updates