Samuel Arzt shows off a project where an AI learns to park a car in a parking lot in a 3D physics simulation.

The simulation was implemented using Unity’s ML-Agents framework (

From the video description:

The AI consists of a deep Neural Network with 3 hidden layers of 128 neurons each. It is trained with the Proximal Policy Optimization (PPO) algorithm, which is a Reinforcement Learning approach.

Dani, a game developer, recently made a game and decided to train an AI to play it.

A couple of weeks ago I made a video “Making a Game in ONE Day (12 Hours)”, and today I’m trying to teach an A.I to play my game!

Basically I’m gonna use Neural Networks to make the A.I learn to play my game.

This is something I’ve always wanted to do, and I’m really happy I finally got around to do it. Some of the biggest inspirations for this is obviously carykh, Jabrils & Codebullet!

In this episode of Visual Studio Toolbox, Arturo Nunez shows us the seamless integration of Visual Studio and Unity and how this makes you a much more productive game developer. You get the benefits of things like IntelliSense and full debugging support for your scripts, as well as Unity specific features like directly implementing Unity API messages in MonoBehavior scripts and the MonoBehavior wizard for adding method definitions.

(And for a limited time, you can take advantage of the Unity Pro and Visual Studio Professional Bundle, which includes Visual Studio Pro, Unity Pro, $50 in monthly Azure credits and more.)


Unity is a game development platform that allows you to create high quality 2D and 3D games, and deploy them across mobile, desktop, VR/AR, gaming consoles or even the web.

In this episode we’re joined by Cloud Developer Advocate Brian Peek (@BrianPeek), who gives us some insight on building games with Unity. He shows how .NET developers can use their skills with Unity and also leverage cloud services in the backend.

Useful Links

Siraj Raval has a video explaining the Unity and it’s new AI tool.

Creating 3D AI in a simulated world is actually pretty easy using Unity. It’s a powerful tool and I’ll go over its new ML Agents toolkit, that allows researchers and developers to build/train ML models in a 3D simulation. It’s pretty fun watching it in a 3D world, lets break down the code and concepts.