LittleWins

LittleWins- Loading Page
LittleWins - Checklist completion pop-up
LittleWins - Registration Page
LittleWins - Achievements & Rewards Page
LittleWins - Initial Dashboard
LittleWins - Final Dashboard
LittleWins - Prometheus Dashboard
LittleWins - Grafana Dashboard
LittleWins - Gemini Analysis Summary

Inspiration

We wanted to solve a simple but frustrating problem: missing a habit log due to bad internet. That small break can ruin motivation and streaks. So instead of building just another habit tracker, we focused on building one that works reliably even in poor network conditions. The goal was simple: your progress should never be lost.

What it does

LittleWins is a habit-tracking Progressive Web App (PWA) that helps users build routines in a simple and engaging way.

What makes it different is how it handles failures. If a user logs a habit while offline, it’s saved locally and synced automatically when the internet is back. On the backend, if MongoDB goes down, the system doesn’t break; instead, it temporarily stores data in a Redis queue and retries later.

We also added an AI-powered SRE assistant that analyzes failures and generates quick post-mortems, so issues can be understood and fixed faster.

How we built it

Frontend: Built as a PWA using React + Vite, with IndexedDB (idb) and Workbox for offline support and caching.

Backend: Developed with Flask (Python), using MongoDB as the main database and Redis as a fallback queue for reliability.

Infrastructure & Monitoring: Containerized using Docker and Docker Compose, with Prometheus for metrics and Grafana for visualization.

AI Integration: Used Gemini 2.5 Flash to generate incident reports and post-mortems.

Challenges we ran into

Keeping data consistent across multiple layers, frontend (IndexedDB), Redis queue, and MongoDB was tricky. We also had to design a retry mechanism that runs in the background without slowing down the main API. Setting up Prometheus and Grafana inside Docker and making everything work together smoothly also took some effort.

Accomplishments that we're proud of

We successfully built a system that doesn’t lose data even when things break. Seeing the app continue working during a database failure, store data in Redis, and later recover everything automatically was a big moment for us.

On the product side, our habit tracker stands out with its F1-inspired theme. We added driver-style radio prompts, streaks, and achievements **to make the experience more **engaging while still keeping the focus on reliability.

What we learned

We learned a lot about building reliable systems, especially offline-first apps, background processing, and monitoring with Prometheus and Grafana. We also explored how AI can help simplify debugging by generating quick and useful incident summaries.

What's next for LittleWins

We plan to add social features, better habit insights, and improve the AI to give personalized suggestions and motivation directly inside the app.