This repository contains materials for an AI Foundation seminar(English version), covering fundamental concepts of AI, Machine Learning, Deep Learning, Natural Language Processing, Transformers with Vibe coding, and Large Language Models (LLMs), including agent-based approaches and related services. It is designed to provide hands-on experience, primarily utilizing Jupyter Notebooks. This is focusing on understanding the machine learning foundation model's concepts, mechanism, code, and development such as MLP, NLP, Transformer and LLM. In reference, you can learn How to develop AI agent with LLM, Computer Vision with Deep Learning and AI for Media Art like below, deeply.

• How to develop AI agent with LLM: This repository contains LLM(large language model), RAG(retrieval augmented generation), AI Agent and MCP(Model Context Protocol) class focusing on creative AI agent development, modeling, and computing as the viewpoint of usecase. The colab code, source, presentation and reference with AI tools like below can be used for developing LLM, RAG and AI Agent.
• Computer Vision with Deep Learning: This course goes beyond simply running pre-existing code. The core objective is to foster a deep understanding by having you implement the internal mechanisms of key deep learning models—such as CNN, ResNet, R-CNN, and YOLO—from the ground up. With hands-on exercises in PyTorch and Keras, you will gain proficiency in translating complex theories into functional code.
• AI for Media Art: This repository includes tutorials and examples to understand how to develop Media Art Work using AI.
  
  

The repository is organized into several folders, each focusing on a specific area of AI, along with supplementary documents:

• 1_AX_trend: AI Transformation trends.
• 2_ML_basic: Basic Machine Learning concepts.
• 3_DL_foundation: Deep Learning foundations.
• 4_NLP: Natural Language Processing.
• 5_transformer: Transformer models.
• 6_LLM_agent_vibe: LLM Agent concepts and vibe coding.
• 7_service: AI services related topics.
• 8_AX_reference: AI Transformation references.
• AI_foundation_and_trend.pdf: PDF slide document possibly detailing AI foundations and trends.
• AI_foundation_syllabus.pdf: The syllabus for the AI Foundation seminar (English version).
• LICENSE: Contains the MIT License information.
• LLM-lesson-plan.pdf: Lesson plan related to Transformer, LLM (English version).
• README.md: This README file.

This section outlines the prerequisites and installation steps (english version) to prepare your working environment for a smooth hands-on experience. All materials can be downloaded from this repository.

First, clone this repository.

git clone https://github.com/mac999/AI_foundation_tutorial.git

Visit the following websites to sign up for accounts. Some services are paid, and it's recommended to set usage limits or subscribe within a certain budget (e.g., $20) for initial experience:

• Colab Pro: https://colab.research.google.com/signup (Paid)
• ChatGPT: (Paid)
• ChatGPT API (Pay as you go): https://platform.openai.com/settings/organization/billing/overview (Paid, set a limit, e.g., $8)
• Claude: https://claude.ai/ (Free)
• GitHub: https://github.com/ (Free)
• GitHub Copilot: https://github.com/features/copilot/plans (Paid, $10/month)
• Hugging Face: https://huggingface.co (Free) - Required for LLM, Transformer, and Stable Diffusion models
• Hugging Face API Token: https://huggingface.co/settings/tokens (Free)
• Stable Diffusion - Kling: https://app.klingai.com/global/membership/membership-plan (Paid, $6.99/month for Standard)
• Figma: https://www.figma.com (Optional)

Note: Record your IDs and Passwords for each account, as they will be used during tool installation.

The following AI services are recommended for project development:

• Gemini API: https://aistudio.google.com/app/apikey (Google AI Key for Gemini)
• SerpAPI: https://serpapi.com/manage-api-key (Google Search API)
• Tavily: https://app.tavily.com/home (Web Search)
• Weights & Biases (wandb): https://docs.wandb.ai/quickstart/ (Model training & fine-tuning monitoring/logging tool)
• LangSmith: https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key (LangChain logging & debug tool)
• Visily AI: https://app.visily.ai (AI planning tool)

Make .env file, input your API key and save it in this repository root folder.

OPENAI_API_KEY=<INPUT YOUR KEY>
HF_TOKEN=<INPUT YOUR KEY>
TAVILY_API_KEY=<INPUT YOUR KEY>
LANGCHAIN_TRACING_V2=false
LANGCHAIN_ENDPOINT=<INPUT YOUR KEY>
LANGCHAIN_API_KEY=<INPUT YOUR KEY>
LANGCHAIN_PROJECT=AGENT TUTORIAL

1. Open colab-env.ipynb from the following link in Google Colab: https://github.com/mac999/LLM-RAG-Agent-Tutorial/tree/main/1-1.prepare
2. Connect to your Google Drive to save practice files.
3. Set up the API keys you created earlier in the "Secrets" menu of your Colab account, as shown in the provided image.

It is recommended to install these tools before the hands-on sessions to save time. Please install stable versions, as the latest versions may cause package installation errors. Ensure you check the "Add to PATH" option during installation if available.

• Python (Recommended 3.12. in 2025.7): https://www.python.org
  • For Mac users, refer to the Python installation guide: https://www.youtube.com
  • Verify installation by running python --version in the terminal.
• NVIDIA Driver (for NVIDIA GPU users): https://www.nvidia.com/Download/index.aspx
  • Skip GPU-related steps if you do not have an NVIDIA GPU.
  • Verify installation by running nvidia-smi in the terminal.
• CUDA Toolkit (for NVIDIA GPU users): https://developer.nvidia.com/cuda-toolkit
  • Check GPU and driver compatibility during installation.
  • Add CUDA path to environment variables.
  • Verify CUDA version by running nvcc -V in the terminal.
• GitHub Tools: https://docs.github.com/ko/desktop/installing-and-authenticating-to-github-desktop/installing-github-desktop
• Anaconda (Recommended 24.0 or higher): https://docs.anaconda.com/anaconda/install/
• PyTorch Library: Visit https://pytorch.org/get-started/locally/ and install the CPU version or the GPU version compatible with your CUDA driver.
• Python Packages (using Anaconda virtual environment):
  • Create a conda virtual environment named venv_lmm with Python 3.12:

    conda create --name venv_lmm python=3.12

  • Activate the environment:

    conda activate venv_lmm

  • Install core packages:

    pip install -r requirements.txt 

• Docker (Optional): Required for container-based operations. Visit https://www.docker.com/get-started/ to install.
• Ollama: Required for local LLM AI tools. Visit https://www.ollama.com/ to install.
• Code Editors & IDEs:
  • Sublime Text: https://www.sublimetext.com/
  • Visual Studio Code (VS Code): https://code.visualstudio.com/download
    • Refer to the provided video for detailed installation and Python extension setup.
    • After VS Code installation, install GitHub Copilot and GitHub Copilot Chat. Refer to the provided videos for guidance.
• Claude Desktop: https://claude.ai/download

Install these if time permits:

• Blender: For LLM-based graphic modeling. Visit https://www.blender.org/download/ to install.
• DaVinci Resolve 20 Public Beta: https://www.blackmagicdesign.com/products/davinciresolve (Optional)

Once the development environment is set up, you can navigate through the Jupyter notebooks (.ipynb files) within the repository's folders (e.g., 1_AX_trend, 2_ML_basic, 3_DL_foundation, etc.) to explore various AI topics and hands-on examples.

This repository is part of my ongoing work on AI, LLMs, and Transformer-based architectures. I am open to research collaboration, academic exchange, and joint projects with universities, public institutions, company and research labs.

For collaboration inquiries, please feel free to reach out: 📧 [[email protected]] | 🌐 [LinkedIn or Personal Website]

Ph.D, Taewook Kang ([email protected])

This project is licensed under the MIT License. See the LICENSE file for details.

For inquiries, please send me email ([email protected]) or refer to the project's GitHub page.