It’s pretty much impossible to interact with tech today without encountering a recommender system, even if you don’t realize it. Think about how many of the platforms you use every day offer suggestions for new content and products, whether it’s your personalized Spotify Discover Weekly playlist, the shows Netflix queues up for you, your social media feeds, or even the potential matches on your dating apps.
Each of these features relies on recommender systems — algorithms that use data to get a sense of your preferences and provide suggestions from a list of options. “Wherever there’s a suggested list of items for us to watch, read, listen to, or buy, there’s a recommender system at work under the hood,” says Nitya Mandyam, Senior Curriculum Developer at Codecademy.
But building a good recommender system can be tricky since our tastes change over time and are influenced by culture and trends. “There’s a mutually reinforcing feedback loop between human and machine behavior that makes recommender systems fascinatingly complex to execute well,” Nitya says. “It’s as much a sociological puzzle as a technical problem to solve.”
Engineers also need to be mindful of the power of recommender systems and consider how the systems they’re building will affect users. “What user behavior is being rewarded here, and is this enhancing or diminishing their quality of life?” Nitya says. “Are there consequences of scaling this recommender system that might cause harm down the line?”
All of this is to say: Recommender systems are an increasingly important skill in the tech job market, and there is a lot of fascinating (and impactful) problem-solving you get to do if you know how to work with them. And now, we’ll show you how in our new Build a Recommender System skill path and Learn Recommender Systems course.
Who are the new courses right for?
As you might expect, knowing how to build recommender systems is a big help for any job that involves using data to make decisions — like machine learning and data science, engineering, or analytics. But according to Nitya, it can also be helpful for people who work in UX research or product design since user behaviors and recommenders are deeply tied to one another. And Front-End and Full-Stack Developers might want to be familiar with recommender systems since they affect how products or items are organized on a website.
Note that while you’ll learn how to build recommender systems in our new courses, professional-grade systems typically use massive datasets that are beyond the scope of these lessons. If you want to work with professional recommender systems, check out our free course, Introduction to Big Data with PySpark, to start learning about how to build things at scale.
What will you learn in the new courses?
If you’re a beginner, a good place to start is our new skill path Build a Recommender System. We’ll walk you through the fundamentals of machine learning and data analysis before you build a recommender system with Python. If you’re already familiar with Python and Pandas, you can jump into our intermediate-level course Learn Recommender Systems to start building right away.
Whether you opt for our beginner’s path or our intermediate course, you’ll learn the differences between recommender system techniques and will understand how to measure the success of a recommender system. You’ll ultimately build systems that provide suggestions for books or movies, but once you learn the basics, you’ll be able to build recommender systems for any domain that’s interesting to you.
“Domain knowledge is key to making a recommender system great, so test out your skills on any topic you’re personally interested in,” Nitya says.
Ready to get started? Sign up today!