Totally Recall

🌪️ Inspiration

Memory loss conditions like dementia and Alzheimer's affect millions of people worldwide, making it difficult for individuals to retain important information or recognize familiar faces. We wanted to build something that empowers individuals to recall memories and strengthen their cognitive abilities using the power of AI. Inspired by loved ones facing memory-related challenges, we created Totally Recall — a compassionate, intelligent, and supportive tool that helps bridge memory gaps with technology. 🧠💡

🛠️ What it does

Totally Recall is an AI-powered assistive app designed to support memory retention and cognitive engagement. It provides:

Daily memory quizzes tailored to the user's history and activities using Gemini AI.
Timely push notifications via Firebase FCM reminding users about important people, places, or past events. a. Notify the user to take a quiz at regular intervals ⏰ b. Send a memory about a person they might be forgetting most 🧓👵
Show random memories to help relive life experiences and aid in better remembering, using FSRS-Anki 📸📝

⚙️ How we built it

AI Core: Gemini 2.0-Flash for tagline analysis and quiz generation. ** FSRS- Anki:** Free spaced repetition scheduler to schedule reviews and notifications at optimal intervals based on learning rate of the person Django, AWS(ECS)- backend: Created a backend REST API with Django and deployed as a server-less architecture in ECS. Flutter Frontend: Created a mobile application using Flutter Firestore: Used Firestore as our database to store quizzes, notifications etc. Auth0: User authentication using Auth0's tokens with Google Sign-In🔐

🔥 Challenges we ran into

Flutter-App: Building a flutter app from scratch was challenge . Especially integrating Auth0 into the application and parsing tokens for authentication made it very intimidating. FSRS: FSRS is a custom python package built on RUST with no proper documentation and developer support. It required manual exploration and research on the algorithm to optimize the parameters.

🏆## Accomplishments that we're proud of Push Notifications: Push notifications serves as a reminder to the user which is crucial especially when they are memory deprived. Real World Impact: This application has immense impact in various industries like Healthcare, Education etc.

📚 What we learned

Custom Package: How to salvage custom packages (fsrs-rs-python) with dependency patching and partial rewrites. Tech Stack Mastery: Leveraging Gemini AI for semantic and tagline analysis, AWS(ECS) for server-less architecture, Django REST API as a backend and a Flutter mobile application. Teamwork at Scale: Coordinating AI/cloud workflows across AWS and Gemini and streamlining the code for such a flow.

🚀 What's next for Totally Recall

Personalization: Personalizing content even more by syncing with photo albums, calendars, and wearable data. 2.Medical Data: Collaborating with medical professionals to test in real caregiving environments. 3.Expansion: Interpolation to other industries like Education