Ember

Full UI View
Reminders
Avatar Selection
One Pager

Problem Statement

According to Y Combinator: “By 2030, 1 in 5 Americans will be over 65, and there’s nowhere near enough people to take care of everyone … 53 million family members are already [caretaking] without pay.” Current solutions like Alexa and Google Home are frustrating to use, can’t hold a stable conversation, and can be unreasonably expensive. Given this, we are convicted that AI has the capability to hold these conversations, and make our users feel more connected and social. Our aim was to take LLMs outside the chat box, and into a more interactive experience. As incoming college students, we know first hand how much love and care our grandparents deserve, so we made Ember to be a personal companion for the elderly.

Solution (Ember)

Ember is an AI companion for the elderly. Simply start a conversation, and an animated 3D avatar will begin a talking with the user. 0 typing required. Ember is highly customizable, with options to change the appearance and voicetone. Ember also has memory capabilities, and can remind users about important tasks (e.g. taking medications, stretching, walking around, etc.).

Jac and Jaseci Lab Usage

JacHammer was the most important program in our tech stack. From the front end UI to the storage of user reminders, to the sound of the voice to the animations, JacHammer handled iterating and designing our app with ease. Specific features we enjoyed integrating were the graph native interfaces and clean API requests.

Lessons and Challenges

Our biggest takeaway was experience. From networking with fellow builders to taking Ember from zero to one, we learned how to ship a product in a timely manner and socialize with a community. While we faced some challenges along the way like accessing more credits in JacHammer and API call requests, we were able to overcome these obstacles by seeking help and thinking outside of the chat box.