Vision, Culture & Leadership
When Keith Gehres arrived as Associate Vice Provost for Enrollment Innovation and Student Experience, he quickly recognized credit evaluation as a top institutional priority. At Carnegie Mellon, each academic college operated its own processes and policies. This decentralized structure meant that students often had very different experiences depending on which program they applied to or which advisor they asked.
“Students had to send the same information multiple places at different points in time, to get ideally the same answer, and sometimes they were getting different answers.”
The need was clear: CMU required an operating model that preserved the autonomy of its colleges while delivering equity, transparency, and reduced workload across the institution.
Having led the world in artificial intelligence education and innovation since the field was created, CMU saw this transformation as part of its identity. The opportunity was not just to fix a broken process, but to demonstrate what a modern, AI-enabled university could look like.
“We love to highlight that we are the birthplace of artificial intelligence, and we are offering the nation's first bachelor's degree in artificial intelligence.”
For Keith, the ultimate goal was seamlessness—a system that just works, without friction for students, faculty, or staff.
“If we do our job well and we do everything exactly the way we are supposed to, nobody knows who we are or what we do… It works the way it is supposed to.”
Partnership with EdVisorly
Keith was introduced to EdVisorly through a trusted friend.
“Probably within the first five minutes, I was like, oh my gosh. Yes. Like, this is exactly the type of thing we’re looking for.”
From the outset, the work went far beyond technology. The EdVisorly team came to campus for multi-day sessions with associate deans, department contacts, advisors, and Registrar staff. Together, the teams mapped existing workflows, clarified policies, and aligned decision makers on what change would require. Only after this groundwork was in place did conversations turn to automation.
“We knew that we had to get the decision makers at the table and excited about this process.”
By prioritizing people and processes first, CMU and EdVisorly built trust across colleges and created a foundation that made technology implementation both faster and more effective.
AP Automation Milestone
Before automation, AP credit posting was fragmented and inefficient. Admissions downloaded scores from the College Board, the Registrar’s Office posted them manually, and advisors sometimes stepped in themselves when timelines slipped.
“Previously, it was redundant, it was inefficient, and it was a little bit of a black box.”
With EddyAI™ and EddyDB™, AP scores are now loaded once, evaluated against CMU’s credit matrix, and automatically applied to student records within the Student Information System. Exceptions are routed clearly to the right stakeholders.
Faculty Collaboration & EddyDB™
Carnegie Mellon’s decentralized structure meant credit decisions often lived in email threads or spreadsheets, with little visibility across colleges. Faculty reviewers, advisors, and Registrar staff worked in silos, and students were left without clarity or consistency.
EddyDB™ has changed the dynamic. Departments still retain academic decision-making authority, but those decisions now flow through a shared system. Instead of chasing information across inboxes, CMU has a central record where outcomes are captured, reused, and trusted. The result is a seamless workflow connecting faculty, advisors, the Registrar, and students.
For faculty, the difference is immediate. In the past, if a student submitted a transcript with a math course, the math department would receive the entire document and have to search line by line. With EddyDB, only the math courses are routed directly into the department’s portal for review. This targeted workflow saves time, reduces confusion, and ensures faster responses for students.
EddyDB also creates institutional memory. Once a course from a sending institution is evaluated, that decision is stored. The next time a student submits the same course, the credit is applied consistently and automatically—building trust and fairness into the process.
By reducing email back-and-forth, routing only the right courses to the right reviewers, and reusing decisions cycle after cycle, EddyDB is creating transparency for students and efficiency for staff.
Implementation & Change Management
Carnegie Mellon knew that fixing credit evaluation was not just a technology project—it was a transformation that required policy alignment, process redesign, and cultural change across a decentralized institution. From the outset, EdVisorly and CMU engaged in deliberate change management, meeting with multiple stakeholders across colleges, the Registrar’s Office, academic departments, systems teams, and enrollment leadership.
The goal was to bring the right voices to the table, build shared ownership of decisions, and map out every step of the workflow before technology entered the conversation. EdVisorly guided these sessions and worked side-by-side with CMU to co-build policies and processes, helping clarify decision-making authority and translate institutional priorities into workflows that could scale. By investing the time upfront, CMU created a foundation where automation could succeed without disrupting academic governance.
“The weekly meetings gave us the chance to lean on EdVisorly’s expertise while shaping our own policies and processes. Together, we could troubleshoot in real time and make sure everything worked hand in hand.”
Weekly working sessions became the engine of progress. In these meetings, stakeholders surfaced policy questions, flagged edge cases, and stress-tested how changes would affect both students and staff. This cadence created an environment of rapid iteration while also respecting the complexity of CMU’s academic structure.
On the technical side, CMU and EdVisorly designed a custom integration into CMU’s homegrown Student Information System. This was no small feat: every new workflow needed to align with existing data models and posting processes, while also creating flexibility for future credit categories such as IB, Cambridge A-Levels, and external transcripts. By combining deep policy work with tailored system integration, CMU avoided the pitfalls of a one-size-fits-all solution and instead built a sustainable operating model for credit evaluation.
This intentional approach balanced CMU’s decentralized culture with the need for central coordination—giving departments clarity, preserving faculty authority, and ensuring the Registrar and systems teams were fully aligned.
Looking Ahead: AI-Enabled Credit Mobility
Carnegie Mellon’s transformation is only beginning. With AP automation and faculty workflows unified in EddyDB, the next horizon is empowering students themselves. CMU will soon enable prospective students to proactively plan their transfer credit before they even apply, bringing unprecedented clarity and equity to the admissions process. By leveraging EdVisorly’s EddyNavigate™ product, the university is turning credit mobility into a student-first experience that reduces uncertainty and accelerates opportunity.
“Just because it has been the way we have done it in the past does not mean it needs to be the way we are moving forward.”
For Keith, this work reflects CMU’s culture of innovation. As the birthplace of artificial intelligence, the university views AI not as a replacement for people, but as a force multiplier, removing friction, standardizing fairness, and allowing staff and faculty to focus on the human side of education.
Carnegie Mellon is proving that vision. By unifying credit evaluation and leaning into AI innovation, the university is setting a new standard for how higher education serves students today and in the future.
