Inspiration
inspired from obsidian banks
What it does
allows your personal agent to search topics about your life and view corresponding connections to offer personal insights, reduced cost for ai to know more about you or the project you are working on current reduction in input cost is 98.8%
How we built it
built using codex gpt 5.5 mostly
Challenges we ran into
insuring the nodes shaped into a natural looking brain, as well as making sure the 3d model spins cleanly with increasing size, improving semantic search for freshly dropped in ai
Accomplishments that we're proud of
making an easy and nice to look at context bank that is agent first, reducing input cost by 98.8 percent for large context projects
What we learned
connections that ai makes should be more broad rather than narrow, for example a project node should relate back to language preferences, art style ect, when the node is related to creating something for a user
What's next for Braincon
whats next for braincon is making it able to run in the background along with your agent taking up minimal ram, and cpu usage for live updates, and context development or changes.
Built With
- codex
- electron
- typescript
Log in or sign up for Devpost to join the conversation.