Course Description
This course explores the evolution of data science programming from traditional code (Software 1.0) to neural networks (Software 2.0) and LLM prompting (Software 3.0). Students will learn to integrate foundational Python tools like NumPy and Pandas with modern AI assistants such as GitHub Copilot and Cursor. The curriculum moves from basic machine learning algorithms to advanced vision and language models. The semester ends with a team-based Kaggle competition where students apply these techniques to real-world datasets.

Course Objectives
By the end of this course, students will be able to:
- Distinguish between the three generations of software and apply them to data science tasks.
- Use AI-powered IDEs and coding agents to improve development speed and debugging accuracy.
- Perform professional data manipulation using the Python ecosystem.
- Build and evaluate machine learning models, including classification, clustering, and deep learning.
- Collaborate in teams to solve competitive data challenges on the Kaggle platform.

• References
o Stanford CS146S - The Modern Software Developer

1 Overview & Logistics
2 Chatting with AI
3 Coding with AI
4 LLM Prompting
5 Modern IDE
6 Coding Agents
7 Python
8 Numpy & Pandas
9 Regression
10 Classification
11 Clustering
12 Dimensionality Reduction
13 Model Selection
14 Interpretable ML
15 Deep Learning
16 PyTorch & JAX
17 Vision Models I
18 Vision Models II
19 Language Models I
20 Language Models II
21 Multimodal Models I
22 Multimodal Models II
23 AI for Data Science
24 Guest Lecture
25 Project Presentation
26 Project Presentation