For Instructors: Google Colab now with auto-grading, powered by Vocareum.

Create a Free Course
Logo wrap
Menu Close

Log in

Request Demo

Bringing AI to Your Campus Interview Series

An interview with Dr. Carl Boettiger

Uncategorized
News
Case Study

Dr. Carl Boettiger

Dr. Carl Boettiger, an Associate Professor of Environmental Science, Policy, and Management at UC Berkeley

In this interview

Welcome back to Bringing AI to Your Campus: A Thought Leadership Series. In this interview, I had the pleasure of speaking with Dr. Carl Boettiger, an Associate Professor at UC Berkeley, where he bridges conservation decision-making, theoretical ecology, and computational methods through data science. He played a role in creating Berkeley’s data science program, growing it from a single freshman course into a comprehensive major, minor, and potentially a new college. Dr. Boettiger’s courses integrate data science and AI into real-world ecological and environmental problem-solving, equipping students to leverage these tools effectively while critically evaluating their limitations and ethical implications.

message-square-01 (1)

AI in the Classroom: Transforming Data Science Education

How has generative AI impacted your teaching?


The advent of tools like ChatGPT has fundamentally changed both how I teach and how students engage with the material. These tools evoke a mix of excitement and concern, particularly around their impact on skill development. While many courses at Berkeley prohibit AI tools, I embrace them as an integral part of learning. In my undergraduate course, Data Science for Global Change Ecology, we use AI to streamline coding tasks, analyze data, and explore new approaches to problem-solving. However, we also delve into its challenges—such as hallucinations, ethical concerns, and environmental costs—ensuring students critically engage with the technology rather than blindly rely on it.

For graduate students, I focus on reproducible and collaborative data science, where AI plays a supportive role in coding and debugging. By integrating tools like retrieval-augmented generation (RAG) with documentation and newer Python libraries such as IBIS and Leafmap, we tackle challenges like outdated model knowledge while empowering students to think critically.

message-square-01 (1)

Balancing Innovation and Foundations

What advice would you give to educators integrating AI into their teaching?


Start by understanding the purpose of your assignments. For example, in data science, the goal is often to develop analysis and problem-solving skills. Using AI to expedite coding or documentation tasks can enhance learning outcomes, but educators must ensure students build foundational skills and know when to drop below the abstraction layer to solve problems independently.

Introduce AI by choosing problems that motivate students, use AI tools to explore solutions, and teach students how to debug when the tools fall short. The classroom environment is critical—it allows students to experiment, fail, and learn how to make these tools work for them. By pairing AI with traditional methods such as manual debugging and documentation, educators can ensure students develop a balanced and practical understanding.

message-square-01 (1)

The Role of Infrastructure in AI Education

What tools and platforms do you use to support your courses?


We rely on Python and Jupyter, using Berkeley’s Data Hubs to provide students with cloud-based environments preloaded with custom packages like Seaborn, IBIS, and Leafmap. Additionally, we’ve partnered with the National Research Platform (NRP) to access cutting-edge AI models on shared GPU clusters. This federated network allows us to host smaller, open-weight language models, avoiding the ethical and privacy concerns associated with commercial platforms while maintaining scalability.

This infrastructure lets us distribute resources efficiently, leveraging server-based solutions instead of assigning each student a GPU. Tools like Jupyter AI enable integration with retrieval-augmented generation (RAG), which allows students to search documentation and refine workflows while exploring the limits of AI capabilities.

message-square-01 (1)

Embracing AI’s Challenges and Opportunities

What are the biggest challenges and opportunities with AI in education?


Generative AI presents incredible opportunities for scaling student learning but also significant challenges around equity, ethics, and technical limitations. AI tools democratize access to advanced resources, enabling students to tackle complex problems. However, they raise concerns about energy use, copyright, and reliance on opaque algorithms.

To address these issues, we prioritize transparency and responsibility in AI use. By teaching students to navigate AI’s limitations and its ethical implications—such as its environmental footprint and potential for misuse—we help them critically evaluate the tools they use. Institutions must support openness, competition, and public academic models to ensure equitable and sustainable AI integration in education.

Similar Case Studies

AWS Academy and Vocareum Partnering to Deliver Hands-On Education to Millions of Learners

AWS Academy and Vocareum Partnering to Deliver Hands-On Education to Millions of Learners

Udacity 1200x600

Udacity

Georgia Tech 1200x600

Georgia Institute of Technology

Platform

Solutions

Company

Resources

Compliance

© 2024 Vocareum, Inc. All Rights Reserved.

Logo wrap

Bring AI to your Campus

© 2024 Vocareum, Inc. All Rights Reserved.