Set the direction
Learners pursue ideas and purposes that matter to them.
About the project
Agentic Coding Studio explores how young people can create with AI-supported coding agents without giving up the judgment, curiosity, and responsibility that make the work theirs.
The opportunity
Coding agents can help translate natural-language goals into computational work. That may lower some barriers to starting—but access to a capable tool is not the same as knowing how to use it well.
Young people still need to decide what is worth making, express intent clearly, judge whether a plan makes sense, test whether a project behaves as intended, and revise it when it does not. The project treats these as central learning practices, not steps to automate away.
Computational agency
Computational agency is the capacity to use computational and AI tools intentionally, critically, and creatively while remaining responsible for goals and decisions.
Learners pursue ideas and purposes that matter to them.
Generated work is something to inspect and critique, not accept automatically.
Testing, debugging, and feedback inform the next decision.
Learners remain accountable for what a project does and how it develops.
A shift in productive struggle
Agentic coding can shift attention toward defining goals, planning, prompting, evaluating, testing, debugging, revising, reflecting, and making design decisions. Technical ideas still matter; learners encounter them through the work of directing and improving a computational system.
Why informal STEM?
Afterschool programs, camps, libraries, community organizations, and creative technology programs can offer flexible, interest-driven contexts for this work.
These settings make room for personally meaningful projects, collaboration, educator coaching, and multiple pathways through a challenge. The project investigates how those conditions shape participation and learning.
Project resource
Explore the motivation, research design, planned development, and collaborators that ground this work.