Foundational 3-month program on AI, Python, and real-world ML applications – developed by TBC Tech School & Geolab
- 📚 Course Overview
- 🧠 Key Topics
- 📂 Repository Structure
- ⚙️ Deve A622 lopment Environment
- 📦 Data Usage
- 🏁 Final Project
- 🖥️ Tmux
- 📜 License
- Type: Foundational Certificate Program
- Duration: 3 Months
- Lectures: 20 (40 hours total)
- Modules: 4 thematic blocks + final project phase
- Instructor: Gurami Keretchashvili
- Instructor-led interactive lectures
- Hands-on coding (Google Colab, Jupyter)
- Individual and group assignments
- Final project presentation
⚠️ Note: Tasks and instructions are in Georgian to reflect the local context of the course.
- Fundamentals of AI, types, and history
- Real-world applications and challenges
- Foundations of machine learning
- Python basics and scripting
- Data structures and file I/O
- Data processing with Pandas, NumPy
- Data visualization using Matplotlib and Seaborn
- Text preprocessing and tokenization
- Vector representations (TF-IDF)
- Sentiment analysis using pre-trained models
- Interactive NLP apps with Streamlit or Gradio
- Image reading and transformation
- Feature extraction and filtering
- Image classification with pretrained models
- Interactive visual AI tools
introduction-to-ai/
├── .git/ # Git version control
├── .gitignore # Git ignore rules
├── assets/ # Icons, banners, visuals
├── flake.lock # Nix flake lock file
├── flake.nix # Nix flake environment config
├── requirements.txt # Python dependencies (alt to flake)
├── README.md # Project documentation
├── module-1/ # Introduction to Artificial Intelligence
├── module-2/ # Programming with Python for AI
├── module-3/ # Natural Language Processing (NLP)
├── module-4/ # Computer Vision
├── tmux.sh # Shell script for tmux-based workspace setup🗒️ Each folder contains Markdown instructions (
task.md) written in Georgian to reflect the local context of the course.
Key dependencies:
- Python 3.11+
- JupyterLab / Google Colab
- Pandas, NumPy, Matplotlib, Seaborn
- Manim (visual explanation tools)
flake.nix support is planned for reproducible development environments.
Since projects vary across modules, data sources and formats will be determined based on the topic and scope of each assignment. Students are encouraged to explore a range of resources including:
- Publicly available datasets
- National statistics
- Crowdsourced or manually collected data
- Web scraping and APIs (where appropriate)
🛠️ Flexibility is key — creativity and initiative in sourcing data are encouraged throughout the course.
The program culminates in a capstone AI project, where students apply acquired knowledge to a real-world problem using data-driven AI techniques. Deliverables include:
- Problem analysis
- Model selection and justification
- Data sourcing and processing
- Visualizations and evaluation metrics
- Live or slide-based presentation
The tmux.sh script sets up a custom tmux session for an organized development environment. It creates a tmux session named "Introduction-To-AI", splits the terminal window into multiple panes for different tasks, and launches relevant programs in each pane. This setup allows the user to code, track tasks, compile and run a programs simultaneously in a single terminal session.
Layout:
_________________________________
| | |
| CODE | |
| NVIM | TASKS |
| | NVIM |
|_____________________| |
| | |
| CONSOLE | |
|_____________________|___________|This repository is maintained by students for educational purposes only. It is not officially affiliated with or endorsed by TBC Tech School or Geolab. All course materials and assignments are based on publicly available or student-generated content.