Rust Fullstack - Ollama: RepStar

This is a submission for the Open Source AI Challenge with pgai and Ollama

progamerxh / rust-rep-star

Rust Rep Star

A Rust project for demonstrating full stack Rust Web.

Table of Contents

• Introduction
• Installation
• Usage
• Contributing
• License

Introduction

This project is designed to showcase how to build full stack Rust Web.

Installation

To install the project, clone the repository and build it using Cargo:

git clone https://github.com/yourusername/rust-rep-star.git
cd rust-rep-star
cargo build

Backend

Run the project using Cargo:

cargo run --bin api-actix

... or in watch mode:

cargo watch -x "run --bin api-actix"

Run the project using Shuttle:

shuttle run --working-directory=api/shuttle

Frontend

Run the project using Dioxus-cli:

cd front/
dx serve

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

View on GitHub

This project was inspired by timescaledb-toolkit repo, which was mostly written in Rust, hence let's build the first Rust fullstack app.

What I Built

RepStar is a comprehensive web application designed to manage and embed testimonials with ease.

• This was my first time diving into the world of Rust for the backend, and I must admit, I was a bit late to the Ollama hype train.
• Nevertheless, the project leverages the power of Rust and Dioxus for the frontend.

Let's explore my code repo for an up-to-date Rust Fullstack app.

Demo

Live demo: https://rep-star-qivi.shuttle.app. The demo showcases the main features of the application, including managing testimonials, RAG summary, Semantic search,...

Tools Used

Backend

• Rust: The core language used for the backend, providing safety and performance.
• Actix-Web: A powerful, pragmatic, and extremely fast web framework for Rust.
• SQLx: An async, pure Rust SQL crate featuring compile-time checked queries without a DSL.
• Tokio: An asynchronous runtime for the Rust programming language, providing the building blocks needed for writing network applications.

Frontend

• Dioxus: A modern, fast, and ergonomic framework for building user interfaces in Rust.
• Tailwind CSS: A utility-first CSS framework for rapidly building custom user interfaces.
• WASM: WebAssembly is used to run Rust code on the web, providing near-native performance.

AI Integration

• Ollama: Used for natural language processing tasks, enhancing the testimonial management experience with features like sentiment analysis and keyword extraction.
• TimescaleDB: Leverages all three of the PostgreSQL extensions for AI: pgvector, pgvectorscale, and pgai provided by Timescale.

Development Tools

• Visual Studio Code: The primary IDE used for development, with configurations for debugging and running the application.
• Shuttle: A deployment platform for Rust applications, used to deploy the backend services.
• Makefile: Used to define tasks for building and running the application, ensuring a smooth development workflow.

Final Thoughts

• Building RepStar has been an exciting journey, combining the power of Rust and the flexibility of Dioxus to create a robust and performant web application.
• Starting with TimescaleDB was incredibly easy thanks to the wizard steps. The available ai and vectorize functions have opened a whole new world for me in PostgreSQL.

Categories: All the Extensions!

• Open-source Models from Ollama: Awarded to a project that utilizes open-source LLMs via Ollama for embedding and/or generation models.
• Vectorizer Vibe: Awarded to a project that leverages the pgai Vectorizer tool for embedding creation in their RAG application.