Voba — AI Recipe Generation

A .NET MAUI app that uses a local Gemma 3:4b model (via Ollama) to generate budget-aware, dietary-safe recipe ideas entirely on localhost.

---

Prerequisites

• .NET 9 SDK
• .NET MAUI workload
• Ollama installed and running
• Gemma 3:4b pulled: ollama pull gemma3:4b

---

Running locally

# 1. Start Ollama
ollama serve

# 2. Clone and restore
git clone https://github.com/YOUR_USERNAME/Voba.git
cd Voba
dotnet restore

# 3. Run
dotnet build -t:Run -f net9.0-windows10.0.19041.0

Connects to Ollama at http://localhost:11434 by default.

---

How it works

1. User sets budget, serving size, lifestyle diet checkboxes, and typed allergies
2. The Interpreter pattern translates restrictions into precise AI rule strings
3. Gemma generates 5 recipe concepts — violations are automatically culled
4. User selects a recipe and gets full step-by-step cooking instructions

---

Project Structure

Voba/
├── AI/
│   ├── Interpreter/
│   └── Pipeline/Handlers/
├── Models/
├── Services/
├── MainPage.xaml(.cs)
└── MauiProgram.cs