You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
2.0 KiB
2.0 KiB
trainable-ai
This is a Vike project. The backend is Hono accepting requests and a tRPC server behind that. The frontend is a React app using Mantine UI for UI and Zustand for state management, but of course with Vike's React integration.
Goal
To create a chatbot that can learn from its interactions with users. The goal is that the user will not have to tell the chatbot to remember anything explicitly; rather, the chatbot will learn from the conversation and remember things on its own.
Project Structure
/assets: Contains static assets such as images and fonts./components: Contains React components/database: Contains modules exporting ready-to-use database clients so server functions can just query the databases without initializing them. Right now, only PostgreSQL is used; but eventually we will use Milvus as well.todoItems.tsis just a toy file./generated: Contains Typescript types generated from the PostgreSQL database schema using Kanel.
/layouts: Contains React layout components intended to be used as generic layouts for pages. Other than that there is nothing special about this directory; it is not treated any differently than any other directory, and is only namedlayoutsfor organizational purposes./pages: Contains the pages of the app, where a file's path is the page's URL path. This is in line with Vike's convention. This is a special directory./server: Contains modules intended to be used on the server-side. There is not yet any particular organization within this directory./trpc: Contains the tRPC code for the server, router/routes, and client./types.ts: Contains the types used by the app. This does not have to be a monolithic file; it can be split into multiple files if it makes sense to do so./state.ts: Contains the Zustand state management code.