Welcome to the world of Cheetah, an innovative macOS app that’s here to revolutionize your remote software engineering interview experience! Imagine having a secret coach sitting beside you, helping you navigate tricky questions and optimizing your performance—all while you engage with the interviewer. Sounds amazing, right? Let’s dive into how Cheetah works and how you can get started!

How Cheetah Works

Cheetah seamlessly blends cutting-edge technology to elevate your interview game. Think of it like a smart navigator guiding you through a dense forest of coding questions:

• Whisper: This tool serves as your ears—working tirelessly to transcribe audio in real-time, ensuring that no crucial detail slips away.
• GPT-4: Picture this as your brain’s assistant, generating hints and answers to promote an insightful conversation. If GPT-4 is unavailable, fret not; GPT-3.5-turbo will step in as a reliable backup!

However, to enjoy these benefits, you will need an OpenAI API key, and it’s best to run Cheetah on a recent M1 or M2 Mac for optimal performance.

Getting Started with Cheetah

Ready to leap into Cheetah? Here’s a step-by-step guide to setting things up:

Prerequisites

• Ensure your macOS is at least version 13.1 or later.
• Check out the whisper.cpp repository in the path ..whisper.cpp.
• Install the SDL2 library by running the following command:

brew install sdl2

Audio Driver Setup

To enhance your audio experience, ensure your input captures both sides of the conversation. If you’re using a video platform like Zoom or Google Meet:

• Utilize BlackHole, a free audio loopback driver.
• Follow instructions for setting up a Multi-Output Device.
• Avoid using the loopback device as the input for your video chat app.

App Overview

Once you’ve set up the app, open it up and select your audio input to begin live transcription. You’ll see a snippet of the transcription displayed under the audio input selector.

Note: Running the app in debug mode may slow down audio transcription significantly.

The user interface features three essential buttons:

• Answer: This generates an answer to the interviewer’s question.
• Refine: Updates the answer based on additional constraints from the interviewer.
• Analyze: Examines your code and logs from the live coding environment in your web browser (requires a browser extension).

Installing the Browser Extension

Currently, Cheetah supports Firefox for the browser extension. Follow these steps to install it:

1. Add the domain of the live coding platform to the matches in .extension/manifest.json.
2. Go to about:debugging.
3. Click on This Firefox.
4. Click Load Temporary Add-on.
5. Select .extension/manifest.json.

Troubleshooting

If you encounter any issues while using Cheetah, here are a few troubleshooting tips:

• Make sure your audio input settings are correctly configured.
• Check your internet connection if real-time transcription is not working.
• Restart the app if you experience lag or performance issues.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

Disclaimer

Cheetah is designed as a satirical art project and not intended for real-world applications. It may produce incorrect or inappropriate solutions, so use it responsibly.

Conclusion

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.