Inspiration
We all have trouble with memory sometimes- especially me (Aneesh). We (Yaseen and Aneesh) built this to help solve this problem.
What it does
MemoryAssist listens in on your environment and detects meaningful moments and conversations and makes a note of them. You can talk to it whenever you want about anything that has happened/you've talked about, and MemoryAssist will be able to help you remember! No more notebooks, no more chugging in context- our app deeply connects with the user over time and understands their pain points of memory!
How we built it
We used claude code (AI development tool) to build this app. This app is mainly an LLM paired with a VectorDB to store memories. Most of the complexity lies in how memories are stored and surfacing the right context when interfacing with the system.
Challenges we ran into
We originally wanted to make this a small piece of hardware that could be attached to your shirt, sleek and smooth. This was hard- we did not have access to the hardware we needed. Additionally, we wanted to use on device ML to ensure security, but again hardware accessibilty was difficult for us.
Accomplishments that we're proud of
We are happy we were able to get our app working and properly understanding the environment context and storing memories well.
Log in or sign up for Devpost to join the conversation.