Building a Game with Amazon CLI in less than 10 minutes.

It’s kind of unbelievable that Generative AI can generate a Multiplayer Game in less than 10 minutes.
(Github Pages Link and Github Repository Link)
With Amazon Q CLI, you can build complex thousands of line of code without actually touching the codebase. With the correct prompt, the game you want to build will be ready without any friction. Amazon Q is now available on cli and also directly on your favourite IDE like VsCode, Eclipse, Jetbrains, Visual Studio etc.

What is amazon Q CLI ?

Q CLI is a generative AI tool which can be used to generate production level applications. Debugging, Refactoring and Reviewing Security of the code has never been this easier. Thanks to Amazon Q. You can set the context for Q, in the project or folder or file level and Q will generate the necessary suggestions, explanations or code for you.

Installation

pip install amazon-q-cli

Choosing the Game

I chose building a simple pingpong game because I always wanted to build one but due to my poor physics and maths skills, I never dared to build one. I know the game looks simple but the maths behind it needs a mature mind.

Building the Game

q cli:
    user:
        "I want a two player game for pingpong."

I couldn’t believe that it actually build very close to what I wanted without telling much.
I had to make few prompts on the colors of paddle and additional features.

Additional Features

On a simple prompt, Q CLI Generated speed boosts and powerups for me. Imagine how many days we would be stuck with brain storming, implementing and testing those features.

Effective Prompting Techniques

To get the best results from AI and avoid hallucinations, I've found a few key prompting techniques incredibly effective. Specificity is paramount – clearly define your goals, desired format, and any constraints. This helps the AI stay focused and reduces the likelihood of it generating false information. Using persona-based prompts (e.g., "Act as a marketing expert...") helps the AI adopt the right tone and perspective, further guiding it towards accurate and relevant outputs. Finally, iterative prompting, where you refine your request based on initial outputs, is crucial for achieving truly polished results and correcting any instances of hallucination that might arise.

Document Generation

It also generated the documentation on a simple prompt

q cli:
    user:
        "now generate a readme file in markdown to summarize all features."

Q Generated Github Action

With a simple prompt you can now generate github actions or your codebuild yamls.

q cli:
    user:
        "can you also generate github actions to deploy to github pages ?"

Gen AI actually wrote a blog for me

Yes I was feeling lazy enough to let gen ai write the blog for me, I even hit the prompt. I was amazed by what it did; but I thought it won’t be able to represent what I truly want to share and the language would be too perfect than what I usually write. I also thought about the temperature issues and wrote the blog myself.

Final Thoughts

Amazon Q Cli is a really powerful tool and I think I’m gonna use it in the long run. I have ben evaluating it since a month and it has been my go to tool since then. It handled so many repetitive tasks so that I could focus on things which Gen AI can’t solve as of now. Overall Productivity has increased in the days. However I’m feeling that I’m already forgetting so many cli commands that I knew and on a complex debugging session, I found myself prompting lot more to Q related to the framework documentation. But since it has increased my productivity, I’m let it stay.

Try it yourself

Give it a go. it has a free version you can evaluate, and it’s available right in your IDE and CLI for a seamless developer experience.

//The information below is posted on 24 Jun 2025 and maynot be relevant at your time of reading
For a limited time, AWS offers new PartyRock users a free trial without the need to provide a credit card or sign up for an AWS account, so that you can begin learning fundamental skills without the worry of incurring costs. Starting 2025, a new and improved daily free usage model will replace the current free trial grant.