🧙‍♂️ Welcome to Not Magic, Just Math!

Welcome, fellow Magicians! ✨

Alright, let’s be real—data science, ML, and AI aren’t really magic (even if they can feel that way). In reality, it is a blend of complex mathematical systems all working together to create results that can seem like magic. But lately, I’ve noticed that non-practitioners are leaning a bit too hard into that magical illusion—selling tools that promise to "magically" solve every problem. That’s not the story I’m here to tell.

✨ The Real Magic-Makers: You 🧙‍♂️

I want to shine a light on the real magic-makers: the people behind the curtain, doing the gritty, often unglamorous, but always fascinating work that makes ML systems tick. No spells, no shortcuts—just solid, tough problem-solving, full of iteration, valuable mistakes, and (yes) tons of math.

🛠 What This Newsletter Is About

This newsletter is all about sharing the stories, techniques, and strategies that help us turn complicated business problems into working ML solutions. We'll dive into how to:

• 🧐 Get to the heart of the problem and frame it for ML

• 🧹 Design and compile a dataset

• 🎯 Choose the right ML approach

• 🛠 Develop POCs that actually convince people

• 🤝 Build buy-in across teams

• 🏗 Design production-ready systems

• 🖥 Select the infrastructure that fits your needs

• 🚀 Deploy your solution successfully

• 🔄 Monitor and improve it over time

In short: the whole process, end-to-end.

🔍 A Peek Into My World

I lead the Data Science team at DLR Group, a large architecture, engineering, and design firm, so you’ll notice a bit of a focus on those areas. But that’s because I find them fascinating! I believe some of the most interesting ML work is happening in places like this—startups and non-tech, legacy industries—where I often find the most skilled end-to-end ML engineers. These people manage the full lifecycle of ML, not just tinkering with a particular part.

📬 What To Expect

My goal with this newsletter is to write weekly on a specific piece of the end-to-end ML process. Sometimes I’ll share my own perspective, and other times I’ll invite friends and colleagues who are deep in this work as well to contribute their insights.

Together, we’ll build a collection of “field notes” on the practice of applied, end-to-end machine learning. As this collection grows, we’ll revisit, refine, and improve on the ideas, creating a valuable resource for both experienced practitioners and aspiring data scientists to explore, get inspired, and start making an impact in their industries.

With that said, if you have a story to tell, or thoughts you want to share from your end-to-end ML work or journey, please reach out by sending an email back to this letter! I’d love to have you contribute!

Sound good? Let’s dig in! 🚀