Issue: Add a "Get Started" Guide for New Users

[CodeRecon-1]

Problem / Motivation
The current repository provides implementations and illustrative code for DeepMind publications, but it lacks an accessible entry point for newcomers. Many users who are interested in learning AI may discover this repo but feel overwhelmed because they don’t know where to start. Without a clear onboarding guide, beginners may struggle to understand how to set up the environment, run examples, or connect the code to the underlying research papers.

Proposed Solution
Add a GET_STARTED.md (or similar) file at the root of the repository that provides:

A simple introduction to the purpose of the repo.

Step-by-step instructions for installation and environment setup.

Guidance on running the first example (e.g., a minimal script or notebook).

Links to key resources (research papers, tutorials, external learning materials).

Suggestions for how beginners can explore the repo incrementally (e.g., start with simpler models before moving to advanced ones).

Impact

Makes the repo more beginner-friendly for students and enthusiasts who want to learn AI.

Encourages wider adoption and engagement from the community.

Provides a smoother onboarding experience, reducing confusion and setup errors.

Aligns with open-source best practices by lowering the barrier to entry.

Additional Context
This would be especially valuable for users who are new to AI research and may not have prior experience with setting up complex environments. A clear “Get Started” file would serve as a roadmap for them to begin experimenting and learning from DeepMind’s research implementations.

[Adrian-patrick]

Hi @CodeRecon-1, I can take this up.
I’ll add a concise GET_STARTED.md focused on environment setup, running a minimal example, and pointers to key papers/resources, keeping it beginner-friendly.

[Sohail-Hafeez]

Hello! I’m Muhammad Sohail Hafeez, a Software Engineering student at NUST MCS with a strong interest in AI.
I’d like to explore contributing a beginner-friendly “Get_Started.md” guide as described in this issue. I plan to first understand the repository structure and existing documentation, then propose a clear onboarding draft for new users.

I’ll share updates or a draft PR once I’ve made progress. Feedback and guidance would be very welcome.

[sahilprajapti-ai]

Hi @CodeRecon-1
solve issue: Add a "Get Started" Guide for New Users
--> Step 1: Create the GET_STARTED.md file

Add a new file named GET_STARTED.md at the root of the repository

Keep it short, simple, and beginner-focused

--> Step 2: Write a clear introduction

Explain in 2–3 paragraphs:

What this repository is

What kind of AI research/code it contains

Who it is for (students, researchers, engineers)

--> Goal: reduce intimidation for first-time users

--> Step 3: Add step-by-step setup instructions

Include:

Python version

Virtual environment setup

Dependency installation (pip, conda, Docker if used)

GPU / CUDA notes (if applicable)

Example structure:

1. Clone the repository
2. Create a virtual environment
3. Install dependencies
4. Verify installation

--> This directly solves beginner setup confusion

--> Step 4: Provide a “first example to run”

Choose the simplest working script or notebook

Show:

Command to run it

Expected output

Explain what the example demonstrates

--> This gives users a quick win

--> Step 5: Add a beginner learning path

Create a section like:

“Where should I start?”

Example:

Start with folder X (basic model)

Then explore folder Y (intermediate)

Advanced research implementations last

--> Helps users navigate large repos

--> Step 6: Link research & learning resources

Add links to:

Original DeepMind papers

Blog posts explaining the models

Beginner-friendly AI/ML resources

--> Bridges the gap between code ↔ research

--> Step 7: Submit a Pull Request

Reference Issue #678 in the PR description

Explain:

Why the guide was added

How it helps new users

Ask for feedback from maintainers

[ionfwsrijan]

Hi there,

I’m Srijan Jaiswal, and I’m eager to contribute to this project as part of my open-source contribution. I have reviewed this issue and feel confident that I can provide a solution since I have experience working with similar technologies in my past projects. Could you please assign this issue to me? I’d love to work on it and contribute to the project.

Best regards,
Srijan Jaiswal