In 1991, Barbara Zimmerman visited the Mẽbêngôkre-Kayapó community of A’Ukre. A’Ukre and Zimmerman came up with an idea to create the Pinkaiti Ecological Research Station (Pinkaiti) within the federally demarcated Kayapó Indigenous Territories in Brazil’s Pará state. Pinkaiti was conceptualized to: (1) preserve Kayapó forests; (2) strengthen Kayapó culture; (3) create an economic alternative to regional mahogany logging; (4) initiate a tropical ecology research program; and (5) strengthen Kayapó transnational networks. After leaving A’Ukre, Zimmerman recruited Conservation International, an international environmental nongovernmental organization (NGO) as an institutional partner. The “Pinkaiti Partnership” has since evolved into a research and education-based multi-stakeholder partnership that includes a transnational network of community, NGO, university, and government actors. Over time, the partnership moved through four eras of activity: initiation (1991-1995); early research (1995-2000); international research (2000-2004); and the field course (2004 – present). Using an embedded comparative case study methodology, this dissertation unpacks the trajectory of stakeholder groups (A’Ukre community, NGOs, universities) as units of analysis to discuss the structure, process, and outcomes of partnership activities across partnership eras. To analyze partnership dynamics, I use Pinkaiti as a boundary object to trace Pinkaiti partner interactions across horizontal, vertical, and transversal axes. As a boundary object, Pinkaiti takes on multiple meanings and forms, depending on its use and context, as it is activated simultaneously or independently by one or more partnership actors. Partnership actors engage one another by navigating cultural, geographic, institution, or knowledge passage points. By tracing each actor group’s trajectory through the lens of Pinkaiti, the study illustrates how boundary objects both permit and restrict transnational collaboration. At the same time, the study reveals both the opportunities and limits of boundary objects as a conceptual tool. Boundary objects can be useful for tracking histories, clarifying the big picture, highlighting feedback loops, and illuminating invisible work. On the other hand, the Pinkaiti study shows that boundary objects can be limited in scope, reflect designer biases, and reinforce unequal power dynamics. Still, the Pinkaiti Partnership suggests important takeaways for actors interested in the design, implementation, or evaluation of education or research-based transnational partnership work.