Automating Data Science Tasks with Generative AI

Generative AI can take the busywork out of data science. This course will teach you how to use generative AI with Python to automate data cleaning, streamline model tuning, and build flexible, agent-driven workflows without relying on a single tool.

What you’ll learn

First, you’ll explore how to automate repetitive and multi-step tasks by chaining prompts to handle data cleaning, validation, and exploratory data analysis (EDA). You’ll see how generative AI can produce pandas-based code to clean messy datasets, summarize key statistics, and surface potential data quality issues-while you remain in control of the process. Next, you’ll discover how generative AI can support modeling workflows. This includes generating baseline pipelines, assisting with hyperparameter tuning and cross-validation, and summarizing model performance.

You’ll also examine how AI can follow common AutoML patterns using libraries such as Auto-sklearn or PyCaret, without requiring deep expertise in every configuration detail. Finally, you’ll learn how AI agents enable more advanced automation strategies. You’ll define what makes an AI agent different from a single prompt and explore how frameworks like LangChain and CrewAI connect generative AI with APIs, databases, and notebooks to automate end-to-end workflows. Just as importantly, you’ll evaluate when agent-based approaches make sense-and when simpler automation is the better choice.

When you’re finished with this course, you’ll have the skills and knowledge of generative AI-driven automation needed to streamline repetitive data science tasks, chain together multi-step workflows, assist with model tuning and AutoML pipelines, and design practical agent-based solutions you can apply in your own projects.

Homepage