This summer, Chuol Ruei Deng, MDevEng ’26, interned at Navajo Technical University, where he helped design a course on AI and Indigenous innovation.

What were the most rewarding experiences you had at NTU?

The most rewarding part was seeing how much more honest and specific AI education becomes when it is designed for a real place rather than a generic student population.

Every session I built had to be grounded in actual Navajo Nation conditions, real water contamination data from abandoned uranium mines, real federal labor statistics mapped against the actual occupational composition of the reservation, the real text of the Navajo Nation Privacy Act read as a living statute rather than a historical document.

That specificity forced a level of rigor I had not fully encountered before. You cannot be vague when the people the course is designed for can fact-check every claim against their own lived experience.

Being on the land itself was also something I did not anticipate affecting me as much as it did.

Visiting sites that have been there for centuries while also knowing that the same land carries the legacy of uranium mining, water hauling, and decades of outside institutions arriving with technology that was never designed with Navajo communities in mind — that combination made the curriculum work feel like something with real stakes rather than an academic exercise.

What did you learn working on this project?

The clearest thing I learned is that Development Engineering requires a discipline that technical training alone does not fully teach. I came in knowing how to evaluate a machine learning model for bias, how to build a data pipeline, how to source and verify research.

What I had not done before was figure out how to make all of that genuinely useful to a student who has never taken a computer science course but who lives next to an abandoned uranium mine, hauls water by truck, and belongs to a sovereign nation with its own legal infrastructure that predates AI entirely. Closing that gap between technical knowledge and community relevant teaching was the real work of this internship.

I also learned something specific about sourcing and accountability that I will carry into everything I do next. Every reading in this course had to be verified, open access, and connected to a specific claim in the session content. Nothing could be included because it sounded right.

I spent significant time confirming the details of the Havasupai Tribe case against Arizona State University, verifying EPA water sampling records against peer reviewed contamination studies, and locating the actual Navajo Nation Privacy Act in the official tribal code to identify exactly where it does and does not protect against AI-specific harms.

A course going before an institutional curriculum committee and eventually before tribal leaders cannot afford to misrepresent a single fact about the community it is designed to serve.

What interested you about the AI and Indigenous Innovation course design?

What drew me to this project was the foundational question underneath it: What does it mean to build technology that serves a community rather than extracts from it?

That question is easy to ask and very hard to answer with any specificity, and this course required us to answer it specifically, week by week, domain by domain. What does extraction look like in health data? In language preservation? In financial technology? In environmental monitoring? In each case the answer is different, and in each case the Navajo Nation has a specific, documented history with outside institutions arriving with tools that were never designed to leave real authority in community hands.

I was also drawn to the structural challenge of building AI education around the Diné Philosophy of Education, Nitsáhákees, Nahat’á, Iiná, and Siih Hasin, in a way that was not decorative.

It would have been easy to label the weeks with Diné terms and teach a standard AI survey course underneath them. What we tried to do instead was let the philosophy actually shape the pedagogy, starting with critical thinking and conceptual grounding, moving through planning and problem identification, spending the longest stretch on implementation and domain application, and closing with reflection and public accountability before tribal leaders.

Whether we succeeded is something the students and community will ultimately evaluate better than I can.