Weaviate Agentic Workflow

AI Dinner Generator with Inngest and Weaviate

This project demonstrates how to build an AI-powered dinner menu generator using [Inngest](https://www.inngest.com/) and [Weaviate](https://weaviate.io/). The application uses RAG (Retrieval-Augmented Generation) to create personalized dinner menus based on cuisine preferences and dietary requirements.

Features

• Cuisine-based recipe suggestions
• Dietary restrictions and allergies handling
• Wine pairing recommendations
• Detailed shopping list generation
• Automatic ingredient substitutions based on preferences

Prerequisites

To run this project, you'll need:

• Node.js installed
• A Weaviate Cloud instance
• OpenAI API key
• Inngest account (optional for local development)

🌱 Setup

1. Clone the repository
2. Create a .env file with the following variables:

OPENAI_API_KEY=your_openai_key
WCD_URL=your_weaviate_cloud_url
WCD_API_KEY=your_weaviate_api_key

📚 Files overview

**`src/inngest/events.ts`**

This file defines the events used to trigger our workflow.

The project comes with the event later used in our example to trigger the Christmas dinner generation workflow.

Events are composed of:

• a mandatory name
• an optional object as data

The events are created with [`zod`](https://zod.dev/) and passed to the Inngest client, enabling an E2E-typed experience.

**`src/inngest/functions.ts`**

This is where our workflow is defined, using `inngest.createFunction()`.

[Inngest Functions](https://www.inngest.com/docs/features/inngest-functions?ref=weaviate-recipes-repo-dinner-generator) are composed of multiple steps, enabling you to write workflows relying on independently retriable steps.

Our first workflow is a simple example of multiple “hello worlds”.

**`src/server.ts`**

Inngest Functions are exposed to the Inngest Dev Server via