People RAG App (using Vercel AI SDK)

People.Rag.App.Demo.mov

A chatbot to store and retrieve information and preferences of your relatives and friends. This intelligent system will only respond with data about people that are present in its knowledge base. You can add information about a relative in a simple chat format and the tool will understand and segregate the information before storing it in the database.

Example Usecase: You can ask what to buy for a friend (whose details are already present in the system) on his/her birthday.

This project uses the following stack:

• Next.js 14 (App Router)
• Vercel AI SDK
• OpenAI
• Drizzle ORM
• Postgres with pgvector
• shadcn-ui and TailwindCSS for styling

Getting Started

Create a postgres database (some possible ways to do so):

• Create a free Postgres database with Vercel Postgres
• Follow this guide to set it up locally

Make a copy of the .env.example file and rename it to .env.

cp .env.example .env

Open the .env file and add

• Postgres connection string
• OpenAI API key.

Install dependencies

pnpm install

Migrate and push database schema to Postgres

pnpm db:migrate
pnpm db:push

Start the development server

pnpm run dev

Start the Drizzle Studio

pnpm db:studio

References

Retrieval-Augmented Generation (RAG) guide.