Project Info
InterSystems: Best Use of GenAI using InterSystems IRIS Vector Search ($2k Cash [1st] & $1.5k Cash [2nd] & $1k Cash [3rd])
MemoryLane
Inspiration
In many developed countries across the world, the population is rapidly aging. This poses a variety of issues to senior citizens, including social isolation, an overburdened healthcare system unable to meet their needs, and the widespread effects of neurodegenerative conditions. We aimed to build a solution which would address all three of these issues in a way which is easily accessible and empowering to senior citizens.
What it does
MemoryLane allows senior citizens to relive and share their cherished memories. The web application combines three main functionalities, which include a journaling and recall feature for important memories, an AI-powered match and chat system for users to discuss their experiences which are shared with other users, and an analytics dashboard which can be used by healthcare professionals to track key indicators of neurodegenerative conditions. Overall, MemoryLane allows users to not only keep their memories fresh but also weave a tapestry of connections with others with similar life experiences.
How we built it
In order to develop a clean and responsive front-end and versatile back-end, we used Reflex.dev to develop entirely in Python. We also used the InterSystems IRIS database to easily perform vector search as well as other database operations to support the backend functionalities required by MemoryLane. Additionally, we made use of the Together.AI inference API to generate embeddings to match users based on shared experiences, perform sentiment analysis to find trends within memory recall data, and to create sample data to test our web app with. Finally, we used Google Cloud to implement speech-to-text functionality to increase ease of access to our platform for senior citizens. The majority of our app was built with Python, with a little JavaScript.
Challenges we ran into
As 2 of our team members had never done full-stack dev before and one was attending his first hackathon, learning the nuances of new frameworks was initially a challenge, especially getting our environments set up. We’re incredibly grateful to the supportive mentors and sponsors for helping us get unstuck when we ran into issues, which indubitably helped us build our final product.
Accomplishments we're proud of
We’re very proud of our clean, intuitive UI which aims to make the product as accessible as possible to our target audience, senior citizens. Additionally, we believe that MemoryLane is a truly unique product which fills a niche which hasn’t been focused on before social media for the elderly, especially in combination with its potential benefits of improving the healthcare industry by aggregating data about the elderly. Also, half of our team was able to go from near-zero web dev knowledge to familiarity with important tools and techniques, which we thought was very representative of the spirit of hackathons – coming together to meet new people and learn new things in a fast-paced creative environment.
What we learned
Our journey with MemoryLane has been an enlightening dive into several new technologies. We harnessed the power of Reflex.dev for frontend and full stack development, explored the nuances in our data with InterSystems IRIS’s vector search on text embeddings from TogetherAI, and learned how to bring text to life with Google Cloud. Together AI has also become our ally in understanding our users' needs and narratives with natural language processing.
What's next
Looking to the horizon, we are definitely looking into expanding MemoryLane’s reach. Our roadmap includes scaling our solution and refining our data model to improve performance, and looking into business models which are sustainable and align with our mission. We envision forming partnerships with healthcare providers, memory care centers, and senior living communities. Integrating IoT could also redefine ease of use for seniors. Keeping innovation in mind, we'll dive deeper into Reflex's capabilities and explore bespoke AI models with Together AI. We aim to improve the technical aspects of our platform as well, including venturing into voice tone analysis to add another layer of emotional intelligence to our app. We believe that MemoryLane is not just a walk in the past – it's a stride into the future of senior healthcare.
Overview
MemoryLane is a groundbreaking web application designed to empower senior citizens by enabling them to relive and share cherished memories. Our platform addresses key issues faced by the aging population, such as social isolation, healthcare challenges, and neurodegenerative conditions, through a blend of technology and personal connection. With MemoryLane, users can journal important memories, connect with others through shared experiences, and provide valuable data for healthcare professionals to track and manage the signs of neurodegenerative conditions.
Features
- Journaling and Memory Recall: A digital space for seniors to record, revisit, and share their life stories and cherished memories.
- AI-Powered Match and Chat System: Leveraging advanced AI algorithms, users can connect and engage in meaningful conversations with others who have similar life experiences.
- Analytics Dashboard for Healthcare Professionals: A comprehensive dashboard offering insights into key indicators of neurodegenerative conditions, aiding in early detection and intervention.
Built With
- Reflex.dev: Utilized for its robust capabilities in full-stack development entirely in Python, ensuring a clean and responsive user interface. *InterSystems IRIS: Employs vector search and other database functionalities to support the backend of MemoryLane.
- Together.AI: Provides AI-driven embeddings for user matching, sentiment analysis, and generation of sample data for testing.
- Google Cloud: Implements speech-to-text functionality to enhance accessibility for users, making the platform more user-friendly for senior citizens.
- Python & JavaScript: The core technologies behind the development of MemoryLane, with a focus on Python for its versatility and ease of use.
Analysis
View
Metric
- 30
- 11
- 7
- 3
Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.
Technology
- JavaScriptIn code
- OpenAIIn code
- PythonIn code
- FastAPIClaimed
3 of 4 appear in the indexed code. 1 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.
AI coding agents
No AI coding agent signals were found in this repository.
Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.
Codebase size
Source size
123 KB
Source files
55
Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.
Repository
gracelliu/TreeHacksX
87 files · 1.1 MB · @ 80501a9
Structure
Interface
21 files · 24%Screens, components and styles rendered to the user.
Application logic
34 files · 39%Domain rules, services and shared utilities.
Supporting
Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.
Languages
- Python91%
- JavaScript8%
- Markdown2%
Share of indexed source by file size. Binary and vendored files are excluded.
Dependencies
dashboard-temp/requirements.txt
pypi · 2- flet
- reflex
requirements.txt
pypi · 2- openai
- reflex
TreeHacksX/requirements.txt
pypi · 1- reflex
Declared in the repository’s manifests at the indexed commit. A declared package is not proof it is used, and runtime dependencies are listed first.
This project’s features have not been analysed yet.
Export this project's context (description, README, evidence, key source files) to chat with an AI agent elsewhere.