The Data Science Education Podcast

“There's no ethics without context.”— Ari EdmundsonIn this episode, we speak with Ari Edmundson of UC Berkeley about his path from European intellectual history to teaching Human Contexts and Ethics of Data, and why understanding the world around a dataset is essential to using it responsibly. He discusses how Berkeley's approach helps students examine the institutions, histories, and relationships that shape technical work. Through a Data 100 project on housing assessments in Cook County, he shows how questions about model accuracy and fairness lead to broader questions about power, public trust, and what happens when numbers enter the world.We also discuss teaching ethics at scale, the role of small-group discussions and graduate student instructors, and the ways students bring HCE into policy work and research. Ari reflects on how the history of the datafied world helps students make sense of generative AI, including debates about automation, work, and AI safety. He shares ideas for expanding contextual reasoning across the curriculum and connecting classic texts to contemporary technology, and closes with a reminder to think of technical work as political work: a way of shaping social relationships that calls for reflection and responsibility.Read the full episode transcript This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

“I think that a focus on mastering the basics without AI is where we should be, especially with undergraduate data science education.”— Bernadette BoscoeIn this episode, we speak with Bernadette Boscoe of Southern Oregon University about her path through art, math, computer science, and information science, and how that background shapes her approach to data science. She discusses using CloudBank and JupyterHub to move students beyond their laptops, giving them experience with shared computing systems and the infrastructures they will encounter in research and professional work. She also reflects on unequal access to computing resources and the need for infrastructure designed around teaching, rather than simply giving students access to commercial platforms.We also discuss machine learning reproducibility, the challenge of preserving a research group’s tacit knowledge, and her work on AquiLLM, an AI assistant that helps researchers navigate the knowledge and materials their teams have accumulated. Bernadette shares how students contribute to Green Crossing AI, a wildlife camera project near Oregon’s Siskiyou Pass, and what they gain from working with real research problems. She closes with her perspective on teaching in the age of AI: building strong foundations, developing critical thinking, and bringing computational thinking into K–12 education.Read the full episode transcript This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“I like to work on intellectually engaging problems, but I also like to maximize for impact.” — Kevin MiaoIn this episode, we speak with Kevin Miao about his path from studying and teaching data science at Berkeley to working on AI and extended reality at Apple, and eventually starting his own company. He shares how his experiences as both a student and instructor shaped the way he thinks about education, and discusses the creation of Berkeley’s post-training course, which was designed to help students connect foundational concepts with rapidly evolving areas like model evaluation, agentic systems, and AI engineering. He also reflects on the challenge of keeping curriculum relevant when technology is moving so quickly, arguing that students should focus less on chasing every new tool and more on learning how new technologies fit into frameworks they already understand.We also discuss the future of AI, XR, and the skills students should develop as technology continues to reshape knowledge work. Kevin shares his perspective on how AI and extended reality could eventually bring digital intelligence more deeply into the physical world, while emphasizing that students should combine technical skills with interests and expertise in areas they genuinely care about. He offers advice on lifelong learning, staying adaptable, and using AI not simply to automate work, but to free people to spend more time on the problems and creative work that matter most to them. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“Just because we have Amazon with Kindle doesn’t mean that we don’t also need community libraries or university libraries.”— Julian Pistorius“Research thrives on openness: open-source software, open data, open weights.” — Javier GuajeIn the first episode of season 12, we speak with Javier Guaje and Julian Pistorius about their different paths into research computing and their work making powerful computing tools more accessible through Jetstream2. Javier shares how his PhD experience led him into research software engineering, where he now helps researchers across different fields. He discusses how open-source software, shared computing resources, and listening to users can support research, along with how AI is changing the tools researchers need.We also hear from Julian, who moved from industry into research computing and found a community built around helping people solve problems together. He shares how hands-on workshops help students and educators build confidence, emphasizing small teams, real-world projects, and learning from one another. Together, Javier and Julian offer practical advice for students interested in the field, from experimenting with new tools to connecting with researchers and developing the communication skills that make collaboration possible. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“Data science shows up in a lot of places where people don’t expect, but at the end of the day, the goal is the same: using data skills and data tools to help organizations make better decisions.”— Mike Alfaro“If it feels hard, it’s because it’s unfamiliar. The more you do it, the easier it will get, and the more fun you’re going to have.”— Annet Isa In this episode of the UC Berkeley Data Science Education Podcast, we speak with recent data science students Mike Alfaro and Annet Isa about their different paths into the field. Mike shares how a data visualization course at Montgomery College first introduced him to the power of storytelling with data, eventually leading him to internships in marketing analytics, transportation, and environmental work. His story highlights how community college, hands-on technical skills, and networking can open doors into data science careers.We also hear from Annet Isa, who returned to school after two decades of professional experience and found data science through her interest in patterns, prediction, and messy data. She discusses a capstone project using GIS, AI, and aerial imagery to identify solar panel installations in Montgomery County, showing how data science can support real-world environmental work. Together, their stories offer practical advice for students beginning their own journeys, from building projects to reaching out to professionals and staying patient through the learning curve. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“Students can say, I understand what’s in this data, because I’m part of the data.” — Alana UnfriedIn this episode, we speak with Alana Unfried, Professor of Statistics at Cal State Monterey Bay, about the future of statistics and data science education. Alana shares her path from classical statistics training to undergraduate teaching, educational research, and her work on MASDER, a national project focused on measuring student motivation, attitudes, and learning environments in statistics and data science classrooms.Alana discusses why data science education needs stronger research tools, better shared data, and a clearer understanding of what students are actually experiencing in the classroom. She explains how MASDER helps faculty collect survey data, compare their classes to national trends, and contribute to a larger picture of what is working across institutions. The conversation also explores major gaps in access to data science education, especially between highly selective and more inclusive schools, and how different departments shape what students learn. Alana also reflects on the growing role of generative AI in data science education and why faculty development will be essential as the field continues to evolve. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“Data science, to me, is all about breaking down walls—breaking down walls between disciplines, and breaking down walls between faculty and students.”In this episode, we speak with Kagba Suaray, Professor of Mathematics and Statistics at Cal State Long Beach, about building a more community-centered vision for data science education. Kagba shares how his work connects data science to local issues in Long Beach and Compton, from public health and housing justice to educational equity, while creating opportunities for students to learn through real, meaningful data. He discusses the power of interdisciplinary collaboration, breaking down barriers that keep students from seeing themselves as “data people,” and designing programs that make data science more inclusive, applied, and community-driven. Kagba also reflects on what it takes to build partnerships, support underrepresented students, and help communities tell their own stories through data. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“A big passion was, how do we think through care? Improvement is fundamentally like a first-order AI problem, not just how to make it easier to do clinical care of today…but how do you make new types of things possible?…If we dig really deep into what’s happening: Why? Why is it caught at this time? Why do we see it in this way? And I think latent into every one of these problems is a frontier AI problem…Through everything—trials and evidence—I think there’s the same type of dynamism we see like in general software, and this kind of pace of change / of improvement that we feel in other parts of AI. Bringing that type of pace to health is the mission of my career, and I’m excited to work on it.”In this episode, we sit down with Adam Yala, Assistant Professor at UC Berkeley and UCSF and co-founder of Voio, to explore how AI is reshaping the future of healthcare. Adam walks through his path from research to building real-world systems, and why computational health is emerging as its own distinct field rather than just an application of AI.We dive into what it actually takes to build in this space, from understanding clinical complexity to navigating challenges like data access and compute. Adam also shares how his experience across academia and startups has shifted his perspective on speed, innovation, and creating meaningful impact.Finally, he offers advice for students and aspiring data scientists, emphasizing the importance of adaptability, curiosity, and focusing on the problems you want to solve in a world where technology is constantly evolving. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“I think we had this feeling that there’s so many students that don’t make it to calculus, and that in the field of data science and the STEM field itself, we really have a gap to fill because we’re missing all of that knowledge and expertise that those students that don’t ever get through calculus would really bring to the field.” In this episode, we speak with Mikahl Banwarth-Kuhn (MBK), Assistant Professor of Mathematics at Cal State East Bay, about reimagining the traditional calculus pathways for today’s data science students. MBK helped develop a new course sequence, Math for Data Science, designed to remove barriers that often prevent students from reaching calculus. She discusses the motivation behind the course and whether traditional pen-and-paper calculus sequences still serve data science students today. MBK advocates for a more intuitive, application-driven approach to help students more deeply understand concepts like derivatives, optimization, and differential equations. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com

Access the full transcript for this episode“When you go out and talk to other people, you realize that you become the opposite of being siloed. You really start to realize that you might have been in an echo chamber when you were talking amongst your own colleagues, and when you start to hear other people, you go, Oh, there’s more that I could understand.”Today, we speak with Rachel Saidi, Professor in the Math, Statistics, and Data Science Department and Data Science Program Director at Montgomery College, a two-year college outside Washington, DC. Rachel shares her path from teaching math to statistics to data science, and what it’s like to scale a data science program in the community college setting, with the goal of catering to students of all ages and experiences. She tackles holistic data science education, combining curriculum, experiential learning, speaker series, and more, while also acknowledging difficulties with constraints like faculty capacity and transfer articulation with four-year universities. Finally, she reflects on how professional organizations can help educators find community and stay on top of best practices, and offers advice to educators and learners on how to tackle data science teaching and learning today. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit datascienceeducation.substack.com