Matt Harrison

Cohort 1

Pandas Masters 1

Cohort 2

Cohort 3

Cohort 4 - April 2023

Cohort 2 - Jan 2025

For Python developers who want to master the modern tools, techniques, and best practices required to create robust, maintainable Python packages. 

Build intelligent, asynchronous systems and leverage OpenAI, Pydantic, MCP, and AI agents.

You’ll go from casually calling LLM APIs to confidently designing, building, and shipping real AI-powered Python systems

AI + Python = Perfection!

Python & Data Trainer to some of the biggest companies in the world

AI Lab

Staff Security Engineer

Data Scientist

Data Engineer

Master the modern tools, techniques, and best practices required for Python

Python Skills for an AI Age

Author of Effective Pandas; 20+ yrs training teams in Python, Data, & ...

Professional Python

Software Developer 

Consultant

Master Pandas patterns to write clean and optimized code, collaborate with others, and prepare for machine learning.

Trainer, Author, Advisor, Consultant

Fundamentals of Data Analysis Using Pandas

Data Specialist 

Data Analyst & Engineer

AI Python Lab is a hands-on, project-driven course for developers who want to go beyond “chatbot demos” and actually ship AI-powered software. You’ll learn how to call any LLM (OpenAI, Anthropic, etc.), enforce structured output with Pydantic, and use coding agents and tools like Aider to dramatically speed up development while staying in control of quality. We’ll connect models to real data and services via MCP, then make everything responsive and scalable using async clients and servers. Along the way you’ll package your work like a real Python project, with clear structure and automation, so you leave the course with a reusable AI-powered application and a workflow you can apply immediately at your job.

What Will Happen in This Course?

Build Real AI-Powered Python Apps

Work Side-by-Side with AI Agents—Without Losing Control

Design Modern, Async, and Automatable AI Systems

You should already be comfortable writing and running basic Python before joining AI Python Lab.

Python

Some familiarity with HTTP APIs and JSON will make things feel more natural, but we’ll review what you need as we start calling LLMs.

APIs

Basic Git/GitHub skills (clone, commit, push) are helpful since we’ll keep code in a repo, though you don’t need to be a Git power user.

Python engineer who wants to move beyond basic API calls and learn how to design robust, production-ready AI services with agents, MCP, and

Data scientist / ML engineer who already prototypes with notebooks and LLMs, but wants structured output, validation, and packaging so their

Technical founder / solo dev who’s comfortable with Python and wants a repeatable workflow to quickly build and ship AI-powered features usi

By the end of this course, students will go from hacking together one-off scripts to building clean, reusable, and testable Python packages they’re proud to ship. Instead of dreading refactors or wondering “did I just break something?”, they’ll have a repeatable workflow: modern environments, clear project structure, meaningful tests, static typing, CI, and a sane way to use AI tools without trusting them blindly. Students will learn how to scaffold a real-world Python project, wire up pytest and mypy, and automate checks with GitHub Actions so every push gives fast, honest feedback. They’ll see how to bring in modern language features—like dataclasses and pattern matching—to simplify code rather than add cleverness for its own sake. Most importantly, they’ll walk away with a working project in a public repo that they can show to their team or manager: “Here’s how I’d set up our next Python service or library.”

From “Copy Paste/AI Hacker” to Confident Modern Python Developer

Projects

Testing

Typing

Continuous Integration with GitHub Actions

That means you can work with functions, modules, and standard library w/o looking everything up.

Basic Python

Cloning a repo, committing, pushing is very helpful

Github Experience

Software Engineer maintaining a growing Python service who wants stronger testing, typing, and CI so refactors stop feeling risky.

Data Scientist who ships notebooks and ad-hoc scripts today and wants to turn them into reliable, reusable Python packages for their team.

Developer who builds Python automation for clients or coworkers and wants a modern workflow that looks professional and is easy to maintain.

I’m Matt Harrison, the author of "Effective XGBoost," a comprehensive guide that has become a cornerstone for data scientists looking to harness the full potential of XGBoost. Over the years, I’ve had the privilege of consulting with numerous clients, guiding them through the intricacies of XGBoost. The insights and practices I’ve shared have often been revelations—unique strategies and methodologies that are not readily available elsewhere. Clients have described working with me as having a personal guide escorting them down a confusing path, transforming their understanding and application of XGBoost from uncertain to confident and effective. In this course, I’m bringing that same level of personalized guidance and depth of knowledge to you. Our curriculum is designed to provide a balanced mix of theoretical knowledge and practical experience. You will learn through engaging video lectures, hands-on exercises, and real-world projects. Plus, with our multi-tiered course offerings, you can choose the level of support and interaction that best fits your learning style and professional needs.

Build Effective XGBoost Models

Course syllabus

Live sessions to enhance comprehension and learn together.

Tuesdays

Each week will have a prerecorded session, homework, and a live session. In addition, you will have the opportunity for peer review and networking.

Six Weeks

Weekly projects allow you to master the content.

Weekly projects

Course schedule

Data Scientists - For seasoned data scientists, this course offers advanced insights and techniques to refine your skills. You will learn c

Machine Learning Engineers - As a machine learning engineer, you will benefit from the deep dive into XGBoost’s functionalities and opti

Software Developers - Software developers interested in expanding their skill set into the realm of machine learning will find this course

Who is this course for

What you’ll get out of this course

Learn advanced Python skills to be productive - lambdas, comprehensions, unpacking, and classes

Master Pandas manipulation techniques - get your data explored and prepared

Leverage scikit-learn pipelines for data prep and feature engineering

Understand model performance - add dollar amounts to models and understand data quality and data quantity

Explore SHAP - Understand your model, constraints, and missing features

Tune to perfection - Use a tried and true method to quickly ensure that your model is not overfitting

Unlock the power of XGBoost and transform your data into actionable insights.
Are you ready to go to the next level?

Learn directly from Matt Harrison

Matt Harrison

About Matt

Courses

AI Lab

Professional Python

Professional XGBoost