Plug and Play: Create a Project G-Assist Plug-in Hackathon
Develop custom plug-ins that add functionality to Project G-Assist, an experimental assistant that runs locally on RTX™ AI PCs. Submit your best entries for a chance to win an NVIDIA GeForce RTX 5090 laptop, NVIDIA GeForce RTX 5080 GPU, or NVIDIA GeForce RTX 5070 GPU.
Hackathon Overview
To join this hackathon, build a custom G-Assist plug-in that adds new commands, connects external tools, and uses AI workflows tailored to specific needs—enabling on-device, AI-assisted functionality that responds to text or voice commands.
The hackathon will run from June 17 to July 16.
Winners will be announced on or around August 20.
See Terms and Conditions.
What is a G-Assist Plug-In?
Plug-ins are custom add-ons that extend G-Assist functionality using its open framework. Some example plug-ins include controlling your PC with voice or text commands, managing peripheral lighting, or summoning other software.
The plug-in architecture makes it easy to get started and define new commands for G-Assist to execute. You can build:
Python plug-ins for rapid development
C++ plug-ins for performance-critical applications
Custom system interactions for hardware and OS automation
Gemini Plug-In
The Gemini plug-in is implemented as a Python-based service that communicates with G-Assist through a pipe-based protocol. The plug-in handles two main types of queries:
Search-based queries using Google Search integration
LLM-based queries using Gemini's language model capabilities
This allows users to leverage the power of cloud AI and search the web without needing to switch programs from the convenience of the NVIDIA App Overlay.
Discord Sample Plug-In
The Discord plug-in is built as a Python-based G-Assist plug-in that communicates with Discord's API. The plug-in follows a command-based architecture where it continuously listens for commands from G-Assist and executes corresponding Discord operations.
This allows users to send messages, charts, and ShadowPlay clips directly to their Discord channels.
IFTTT Sample Plug-In
The IFTTT plug-in is built as a Python-based G-Assist plug-in that communicates with IFTTT's API. The plug-in follows a command-based architecture where it continuously listens for commands from G-Assist and executes corresponding IFTTT operations.
This allows users to trigger IFTTT applets and manage smart home devices using voice commands.
Hackathon Process
Step 1: Start Now
Register to join the hackathon, and get a head start with our curated technical resources.
Step 2: Build Your Project
Get started creating an amazing plug-in!
Explore our G-Assist repository and try out our ChatGPT plug-in builder assistant.
Join the NVIDIA Developer Discord channel to connect with NVIDIA experts and other developers.
Step 3: Submit Your Entry
Submit your project with the following:
GitHub repo including:
Source Code File (plugin.py)
Requirements.txt
Manifest.json
Config.json (if applicable)
Plug-in Executable File
READme
A short video highlighting the plugin in action (0:30–2:00)
Social post promoting your plug-in and video overview on X, Instagram and/or TikTok, using #AIonRTXHackathon
Getting Started Resources
G-Assist GitHub
NVIDIA’s GitHub repository provides everything needed to get started on developing with G-Assist—including sample plug-ins, step-by-step instructions, and documentation for building custom functionalities.
ChatGPT Plugin Builder
Transform your ideas into functional G-Assist plug-ins with minimal coding. This tool uses OpenAI’s Custom GPT to generate plug-in code, making it easier than ever to extend G-Assist’s capabilities. Whether you want to create a weather plug-in, a task manager, or any other custom functionality, the Plugin Builder streamlines the entire development process.
Create a Plugin with Python
Transform your Python applications into powerful AI-enabled experiences with G-Assist! Our G-Assist Python Binding and instructions makes it incredibly easy to integrate G-Assist's capabilities into your Python projects. We've abstracted away the complexity of state machines and callbacks, making everything beautifully synchronous and straightforward.
How to Create a Plug-in Walkthrough
In this blog, we’ll walk through the architecture of a G-Assist plug-in using a Twitch integration as our example. You’ll learn how plug-ins work, how they communicate with G-Assist, and how to build your own from scratch. Whether you’re a Python dev, C++ enthusiast, or just getting started, we’ve got tools and templates to make plug-in development fast and approachable.
Judging Criteria
Innovation and Creativity
Show us your most innovative plug-in ideas.
We can't wait to see what you dream up!
Technical Execution and Integration
System Check: All Green
Looking at how solid the foundation is—technical depth, G-Assist integration, scalability, and how well the tools were put to work.
Usability and Community Impact
We’re looking for easy to use plug-ins that can help anyone with an NVIDIA RTX GPU.
Prizing
First Place
RTX 5090 Laptop
DLI Self-Paced Course Credit
Social Promotion by NVIDIA
Meeting With NVIDIA G-Assist Team
Second Place
RTX 5080 GPU
DLI Self-Paced Course Credit
Social Promotion by NVIDIA
Meeting With NVIDIA G-Assist Team
Third Place
RTX 5070 GPU
DLI Self-Paced Course Credit
Social Promotion by NVIDIA
Meeting With NVIDIA G-Assist Team
NVIDIA Developer Discord
Need help as you build your plug-in? Join our Discord to share your creations, join G-Assist conversations, and get help from our community or from NVIDIA experts directly.
RTX AI Workshop: How to Build a G-Assist Plug-in
Need more help or have questions? Join us for our live webinar on June 24th at 10AM PT where we will walk through how to create and customize G-Assist’s functionality, add new commands, connect external tools, and build AI workflows tailored to specific needs—enabling quick, AI-assisted functionality that responds to text and voice commands.
You’ll have an opportunity to ask our team questions about your projects and share any roadblocks during a live Q/A.
Ready to build and share your own G-Assist plug-in? Join the hackathon now.