Project Info
This project did not submit a demo video on Devpost.
Inspiration
Inspired by our personal experiences with older relatives and grandparents, we were struck by the surprising absence of voice-automated tools designed to support their daily lives. In response, we created Yaad—the Hindi word for "remember"—as a means to unlock and preserve memories through intuitive recitation and retrieval. We believe this approach not only enhances accessibility but also offers a dignified way to maintain personal connections, and we hope you will share in our vision.
What it does
Yaad takes in input for two different functions: storing memories and retrieving memories. The user can select what they want to do and speak to the interface and get tasks and memories retrieved for them based on relevance.
How we built it
Yaad was built using Deepgram, LangChain, Groq, OpenAI, React.js, and TypeScript
Challenges we ran into
Due to the scope of our project, our most difficult issues were with merging the backend logic with our frontend UI.
Accomplishments we're proud of
This is the first time that team Yaad has worked together on a Hackathon project, and it was definitely a great experience with our complimentary skills in Full-Stack, Backed, and UI/UX development.
What we learned
In this project, we learned so much involving expressing information in new and creative ways, from generating vector embeddings in Pinecone DB to interpreting and displaying those embeddings through RAG modeling.
What's next
We want to expand our feature set to image classification and interpretation, as well as increase the amount of accessibility features to improve our target audience's quality of life.
Yaad
Built for the 2024 CalHacks Competition
Mission
Our mission is to create a "living memory" database that preserves and recalls cherished memories for individuals with degenerative cognitive conditions, such as Alzheimer's and dementia. By integrating AI and advanced vector-mapping technology, we offer a meaningful way to record, transcribe, and store memories, making them easily accessible through conversational prompts. Our goal is to enable users to relive their past experiences and create moments of joy, familiarity, and comfort. We aim to foster a deeper sense of connection and provide a dignified, supportive tool that nurtures memory and personal identity.
About
Yaad is an innovative web application designed to preserve and revisit the cherished memories of individuals, particularly those with degenerative cognitive conditions like Alzheimer's and dementia. Our app allows users to record, transcribe, and store their memories securely while using AI to retrieve specific experiences based on prompts. With this unique blend of technology and empathy, Yaad aims to foster moments of connection, comfort, and joy through personalized memory recall.
Technologies We Use
Next.js
Our app is built using Next.js, a powerful React-based framework that enhances performance and enables seamless server-side rendering. This ensures a smooth user experience, with rapid page loading and efficient memory recall, no matter where users are located.
Clerk for Authentication
We use Clerk to manage user authentication securely. It provides a user-friendly login and sign-up process, supporting features like passwordless login and multi-factor authentication, ensuring that user data and memories are kept safe and accessible only to authorized individuals.
Pinecone for Database and Vector Embeddings
At the core of our memory retrieval system is Pinecone, a high-performance vector database solution. Pinecone allows us to store vector embeddings of recorded memories, enabling our AI to quickly and accurately retrieve related memories based on specific prompts. This advanced search capability makes it easier to find and relive memories that may have faded over time.
Tailwind CSS for User Interface
We’ve designed the Living Memory interface with Tailwind CSS, a utility-first CSS framework. Tailwind enables rapid, responsive UI development, ensuring that our app is visually appealing, accessible, and easy to navigate for users of all ages and tech familiarity levels.
Team Yaad
- Aarush Jagtap – Full Stack Engineer
- Pranav Singh – Back End Engineer
- Siddhant Bhardwaj – UI/UX Engineer
Analysis
View
Metric
- 47
- 30
- 4
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
- CSSIn code
- LangChainIn code
- Next.jsIn code
- OpenAIIn code
- ReactIn code
- Tailwind CSSIn code
- TypeScriptIn code
- JavaScriptClaimed
7 of 8 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
79 KB
Source files
35
Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.
Repository
ajagtapdev/yaad
57 files · 10.1 MB · @ 20496cf
Structure
Interface
26 files · 46%Screens, components and styles rendered to the user.
API & routing
4 files · 7%Request entry points: routes, handlers and controllers.
Application logic
4 files · 7%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
- TypeScript93%
- CSS4%
- Markdown3%
Share of indexed source by file size. Binary and vendored files are excluded.
Dependencies
package.json
npm · 36- @clerk/nextjs
- @clerk/themes
- @deepgram/sdk
- @langchain/pinecone
- @pinecone-database/pinecone
- @radix-ui/react-separator
- @radix-ui/react-slider
- @radix-ui/react-slot
- @singlestore/client
- class-variance-authority
- clsx
- cross-fetch
- dotenv
- emailjs-com
- framer-motion
- fs
- groq-sdk
- langchain
- +18 more
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.