Empowering the Innovators of Tomorrow: Why Practical Machine Learning Education Matters

Machine Learning (ML) is shaping the future of the next generation of thinkers, builders, and decision makers. And the quality of their education today will define how responsibly and effectively they use ML tomorrow. That is where the BigML Education Program steps in. With more than 850 prestigious universities and schools worldwide already on board and 12,000+ students trained since 2016, BigML is the go-to learning platform for educators preparing students to use ML to solve real-world problems.

The Importance of High-Quality ML Education

Every organization, whether it’s a startup, hospital, logistics firm, or government agency, is recording data that can be key to making the right decisions. That data is constantly growing while decision makers are trying to find the right strategies for success. Machine learning can provide the bridge to access the information that our data contains, but you need some real training to cross that bridge. Therefore, leaders and professionals don’t just need to understand machine learning; they also need practical experience to apply it to improve decision-making, automate workflows, and approach new opportunities. Theoretical-only courses alone will not get them there. They need tools that let them build, test, and collaborate in a no-code, human-friendly environment. That is where the BigML platform becomes very useful for educational purposes, as it is intuitive enough for beginners, powerful enough for advanced projects, and definitely very helpful for educational matters.

Why Educators Choose BigML

The BigML platform is beginner-friendly and robust. Its visual interface gets students up and running quickly. Meanwhile, advanced users can dive into its full REST API, automation tools like WhizzML, its domain-specific language for automating machine learning workflows, as well as a deep library of ML resources.

Educators love the BigML Organizations feature, which gives them complete classroom control. They can:

• Decide who is enrolled per term
• Create and manage collaborative projects
• Grant access rights by project or group
• Reorganize groups or preserve student work across terms

This flexibility allows instructors to decide how students work, collaboratively or individually, depending on the course structure and goals. The following video presents the benefits of using BigML Organizations in the classroom, from the instructor’s perspective.

Students can also benefit from using the BigML Organization. Their work is constantly available to teachers and teammates, who can track their progress and provide help at all times. This video also tackles the students’ perspective. 

Meet some Instructors Using BigML in their Classroom!

From Sweden to Spain, Switzerland to the U.S., educators across the globe are using BigML to teach machine learning in a way that sticks. Below are four instructors presenting their testimonials on how and why they use BigML Organizations to bring ML to life in their classrooms.

As a post-doc back in 2021, I started to do research on AI in organizations. It became a natural step to integrate this research with teaching courses on AI. I think ML is a fascinating technology that requires a diverse set of skills to be successfully implemented, such as analytical capacity, creativity, and critical thinking.

Additionally, I believe in a “learning-by-doing” approach to teaching technology. It is very difficult to learn central concepts such as “human-in-the-loop”, “bias,” and so on without seeing technology in action. The risk is always that the teaching becomes too general and abstract.

My goal is to give students from non-technical oriented disciplines hands-on experience of the opportunities and challenges associated with using ML to solve business cases. For that, I usually compare BigML to a Swiss knife: it is versatile, and allows for a range of data sets to be used to train and evaluate a variety of ML models.

As for how useful the BigML Organizations are for me, I can say they are central to my teaching approach. I usually divide the students into groups where they are to solve business cases. BigML Organizations help a lot when it comes to user management, and it is central for being able to supervise the students in their work.

On top of that, integrating the BigML tool into my existing curriculum was quite easy, as the threshold to start using BigML is fairly low, which makes it even more convenient.  I went through the BigML analytics training program, which was beneficial to show some additional features, but overall, I think the functionality in the platform is pretty self-explanatory.

Regarding the impact on teaching and learning, my students find BigML easy to use, and they enjoy being able to train ML models without coding, which is a great advantage in the teaching context, along with the BigML Organizations. As BigML automates many parts of the ML workflow, we have received positive feedback from the students about how they can spend time working on the data, rather than coding, and they can work iteratively and be creative and try different things in the platform.

For instance, since the platform enables rapid experimentation, the students can engage in iterative training of ML models, and see what happens if they make some changes in the dataset, etc. In my classes, I put a lot of focus on the evaluation of models, and I think the ROC curves in the platform are a good way to illustrate the tradeoffs (between precision and recall) and choices that are necessary during ML development.

When it comes to real-world preparation, in combination with our live-case approach, where we give the students organizational cases to solve with BigML, the usage of this ML platform gives my students insights into the complexities surrounding ML development.

I would also like to share that the research and teaching team in Umeå was nominated for the ECIE teaching awards for our use of no-code platforms such as BigML in higher education. We have also written a couple of academic papers on the topic, where we specifically mention BigML:

• Sundberg, L., & Holmström, J. (2024). Teaching Tip: Using No-Code AI to Teach Machine Learning in Higher Education. Journal of Information Systems Education, 35(1), 56-66.
• Sundberg, L., & Holmström, J. (2023). Democratizing artificial intelligence: How no-code AI can leverage machine learning operations. Business Horizons, 66(6), 777-788.

Finally, I would call out other educators to consider BigML in their ML courses for everything explained above. I am currently running a research project on the use of no-code AI in higher education, so don’t hesitate to reach out if you want inspiration / exchange experiences of using platforms like BigML! I think no-code AI tools like BigML will be a natural part of courses in business intelligence / analytics in the future.

My journey into teaching Machine Learning began unexpectedly in 2017 when ESADE Business School in Barcelona reached out for a guest lecture. This initial experience ignited my interest in teaching practical ML, and soon, other business schools also requested guest lectures from me. Therefore, I decided to teach ML because there is a clear demand for ML literacy in the business world, and I aim to meet that need. By teaching ML, I’m helping shape future business leaders who will seamlessly integrate this technology into their strategies. Currently, I use BigML for teaching at the University of Fribourg, the Bern University of Applied Sciences, and during my guest lectures at ESADE Business School in Barcelona. 

I chose BigML to teach because it is the only practical option for me, thanks to its intuitive web interface which does not require coding. It has been clear to me from the start that expecting non-technical students to code would be the wrong educational approach. 

BigML’s very intuitive way to build end-to-end Machine Learning workflows also makes it very convenient to explain  ML concepts. The web interface enables me to tell stories and illustrate key concepts seamlessly during my teaching. One of my favorite exercises in BigML is demonstrating the unbalanced class problem while deepening the understanding of evaluation metrics like recall and precision. It’s a powerful way to connect theory with practical insights. Also, I like to see the reaction of my students when they practise with BigML, which in most cases, they are focused and quiet with a subtle smile at the corner of their mouth.

As for the next steps with my students after thy finish their education, many of them often choose BigML for their program degree theses, which proves they like and enjoy the tool. The objective in my courses is not to become a data scientist but to learn how to work effectively with data scientists without getting intimidated by the data science speak. This works very well.

Generally speaking, the feedback from my students about the BigML platform is very positive, especially if they have had a previous experience with other less intuitive Machine Learning tools. For me, as an ML educator, it’s a no-brainer, my role is to eliminate unnecessary barriers that hinder students from grasping fundamental concepts and to spark their creativity to experiment and explore. Achieving this requires a user-centric machine learning tool, and BigML is the best one I’ve encountered for this purpose! By lowering the entry barrier, it helps democratize Machine Learning, making it more accessible to non-technical domain experts, key contributors to the success of AI-powered solutions within their organizations. What’s particularly remarkable is that the great user experience doesn’t come at the cost of depth or rigor. Instead, it is often the result of great engineering, which —though perhaps a controversial take— could serve as a valuable lesson for some data scientists who underestimate the importance of well-designed, structured Machine Learning workflows.

I used to work at a company that developed ML models for customers, so I was used to working with data and running ML projects. Now, as an ML instructor, I think that everybody should know how AI works, and if you want to understand the world, you should learn ML, as ML is going to be a key tool for many companies and individuals. 

Adopting BigML in my classroom was smooth and simple. BigML is a very easy-to-use, intuitive, and highly visual tool. Students can navigate through decision trees, view the data, and analyze it with histograms, scatterplot visualizations, and more. I believe the strength of BigML lies in its highly visual nature while also being powerful for automating processes and implementing models in production. In this way, it is useful for teaching both technical students and advanced programmers.

On my personal side, I obtained the BigML Engineer and BigML Architect certifications, which helped me a lot in learning more about BigML and also ML concepts, which I integrate into my classes. However, I like to start with theoretical concepts (what is ML, how it works, what data we need, in what format) and then apply those concepts using BigML. I usually begin with a simple dataset, review it on my computer (rows, columns), upload it to BigML, and start playing around with it.

The truth is that my students find BigML very easy and intuitive! I always tell them that it might look like a toy, but in fact it is a powerful tool. It’s not only for learning, but also for developing and deploying real-world applications. There are specific features that are very popular among my students, for instance, the 1-click-everything is magic! Abstracting from what is happening under the hood is key to letting students concentrate on the results. In my experience, although ML is basically statistics, with BigML, you don’t need to teach a single math formula, and this is very helpful! Also, the decision tree visualization is key to understanding what a pattern is. The fact of being able to go through the branches, splits and leaves, with the mouse, while you see the data, lets the student understand that a pattern is made up of similar datapoints.

I think that the tool helps prepare students for real-world applications of Machine Learning to learn all the concepts, from training to the evaluation process. The BigML platform allows students to have all the data and resources in the same place. I usually say that BigML auto-document projects, as you don’t have to keep track of what CSV file you used to train this or that model, or what their parameters were. Everything is in the platforms, and you keep traceability of all the steps.

To close my testimonial, I would like to encourage ML educators worldwide to try BigML in their classroom and see how easy it is to teach any ML concept with it. I have used some other tools and have never seen one that has as many features as BigML, including computer vision!

BigML is not only used in universities, we go beyond any barrier and are happy to help any educational institution that chooses for good-quality education. That is the case of our last testimonial of this blog post, Helen Burn, an instructor at Highline College, which recently started using BigML thanks to their partnership with the University of Washington, where they also use BigML to teach ML! 

Highline College is a public college in Washington state, located midway between Seattle and Tacoma. Originally a community college, we now offer seven Applied Bachelor’s Degrees. In 2023-24, we served over 15,000 students. Our student body primarily draws from South King County, a region known for its rich cultural diversity, creating a dynamic and vibrant learning environment. Approximately 30% of our students are pursuing transfer degrees, while 16% are enrolled in professional technical programs. The largest group, accounting for 41%, is focused on high school completion or basic skills education. 

In spring of 2023, staff from the eSciences Institution at the University of Washington and I connected at the Academic Data Science Alliance Leadership Institute and over the next few months discussed expanding data science offerings to first and second-year college students at institutions overseen by the State Board of Community and Technical Colleges. Highline College is taking a leading role in this collaboration. Their Introduction to Data Science course, which we offer in the winter term of 2025, includes a unit on machine learning. The partnership with the University of Washington has launched and supported an initiative across the state aimed at incorporating more data science curriculum into the first two years of college. Introduction to Data Science (CSCI 180 at Highline College) is a pilot “non-coding” introductory course designed as a survey class that other colleges can adopt. The University of Washington currently offers a data science minor, and by integrating this curriculum into the SBCTC system, our goal is to better position students who start their academic journey at SBCTC colleges to complete the UW data science minor. Additionally, this course helps prepare students for data-intensive coursework they will encounter as juniors and seniors at their transfer institutions. It also serves as a springboard for students interested in pursuing data science majors, which are rapidly growing across the state. Introduction to Data Science (CSCI 180) emphasizes hands-on activities and projects, teaching data science skills through user-friendly commercial tools like BigML. This approach makes the course equally accessible to students in data-adjacent fields or those looking to enhance their data competencies.

Every student deserves the opportunity to understand how data drives artificial intelligence. A truly authentic and modern introduction to data science must include some discussion of machine learning.

Finally, it is important to note that the primary challenge when teaching Machine Learning lies in the fact that the most popular Introduction to Data Science curricula typically rely on programming languages like Python or R. While I won’t name specific examples, these approaches often overwhelm students by requiring them to develop competencies in three areas simultaneously: data science, statistics, and computer programming. BigML, with its menu-driven interface and user-friendly design tailored for professionals seeking less technical tools, offers a more accessible pathway. It enables students to focus on understanding the core concepts of how data science is applied in machine learning, promoting greater success.

We chose BigML because it is also used by the University of Washington, and we share the same motivation outlined above: to provide students with accessible tools that allow them to focus on understanding the core concepts of how data science is applied in machine learning, without the added burden of learning a programming language.

The Global Movement in ML Education

Highline College and the University of Washington are just two examples where BigML is the preferred tool to teach Machine Learning in the classroom in the US, but there are hundreds! This map of our Education Program page presents the more than 850 educational institutions worldwide that choose BigML for educational purposes, such as the United Nations System Staff College (UNSSC), which has been using BigML for several years now and shares it with the world through LinkedIn posts or with interviews with their staff.

These stories shared today in this blog post are just a sample of a much bigger picture, and there will be more testimonials coming soon. Stay tuned! From business schools, tech degrees, and high schools to graduate programs, educators worldwide are choosing BigML because it is practical, scalable, and puts real ML into the hands of students, no matter their background. Thanks to this, thousands of collaborative ML projects have been launched over almost a decade.

If you are an educator and you want your students to do more than just watch lectures, e.g., you want them to build, experiment, and solve real problems, it is time to take the next step! Join the BigML Education Program by contacting our Education Team at education@bigml.com and requesting access to the BigML Organizations for your institution.

Remember that the future of Machine Learning deserves the best education we can give it, and it starts in your classroom!