Research

The Architecture of Becoming: Designing Career Development Systems as Longitudinal Learning Environments (2026)

The paper argues that most career platforms are built for the institution, not the student. They prioritize placement metrics and operational efficiency while ignoring the reflective work students need to do. Students end up bouncing between disconnected tools — resume builders, skill inventories, advising systems, networking platforms — each showing a different version of them, with no way to connect the pieces over time.

We propose a framework with four parts: reflection (structured writing and exercises), interpretation (tools that help you spot patterns without telling you what they mean), visualization (maps and diagrams that let you see your own thinking), and action (job search, mentorship, and skill-building that come after you understand yourself, not before). Read the full paper.

Think first, act second. Students who spend time on their own patterns and priorities before jumping to job listings, alumni networks, or industry research consistently make better decisions.

The best data about a student comes from the student — what they write, what they say, what they notice about themselves through structured reflection. Not what the institution tracks about them.

Our tools show you what might be there — they don't tell you what it means. Students work through what they're seeing in conversation with peers and facilitators, not by reading a printout.

Career maps, pathway diagrams, and spatial layouts help students see connections they can't hold in their heads. When you can look at your own thinking from the outside, you notice things you'd otherwise miss.

Student growth doesn't happen in isolation. We build tools that connect what students, advisors, and employers see — because development is an ongoing conversation between a student and the world they're entering.

98 pilot participants. Self-reported clarity shifted from 2.3 to 4.2 (5-point scale). 90+ journal entries across academic terms. 70+ themes identified through qualitative coding of onboarding responses and journal entries.

60+ workshop maps created across cohorts. Clarity shifted from 2.1 to 4.4. 85% of participants reported a clearer personal story after the exercise. 100% named at least one insight they hadn't put into words before.

500+ students tested across institutions. Six either/or choices across three dimensions (Values, Strengths, Skills) produce 64 possible anchors. Students who got a single word had more substantive peer conversations than students who got detailed profiles from longer assessments.

A 90-minute facilitated workshop where students compress their story from a paragraph to a sentence to a single word, then sort cards across values, skills, and strengths. Built in response to a NACE finding: 78% of students say they communicate well, but only 54% of employers agree. The issue isn't missing experience — it's that students haven't learned to tell their story.

Every project starts as a prototype. We deploy it with real students, watch what happens, collect data, and use what we learn to build the next version. Each tool is a hypothesis about what helps people think — and each academic term is a chance to test it. We treat career development as a design problem, not a service delivery problem.

We combine computational text analysis with qualitative research. On the qualitative side: student writing, facilitated conversation transcripts, spatial arrangement documentation, and journal entries collected over time. On the quantitative side: engagement data from digital tools (Synapse, DartWorld) and pre/post measures of how clearly students can articulate their story and how confident they are in their decisions.

Everything gets tested with real people in real settings — not simulated in a lab. Our primary research site is Dartmouth College's Center for Career Design. We also teach and run programs at Kent State University.

The work draws on learning sciences (reflection, metacognition, distributed cognition), human-computer interaction (external representations, cognitive load), identity development theory, narrative psychology, and sensemaking research. We treat career development as a form of learning — not a transaction — which means it needs environments that support sustained reflection, pattern recognition, and storytelling over time.

Baxter Magolda (2001) on self-authorship and identity development. Bransford, Brown, and Cocking (2000) on how people learn. Schon (1983) on reflective practice. Weick (1995) on sensemaking in organizations. Mezirow (1997) on transformative learning. Kolb on experiential learning cycles. Norman (1993) and Kirsh (2010) on external cognitive supports. Zimmerman (2002) on self-regulated learning. Card, Mackinlay, and Shneiderman (1999) on information visualization.

Comparing student clarity and goal stability across different tools over multiple terms. Do the gains from structured reflection last after the program ends?

What does it take to move an institution from counting placements and job board clicks to measuring things that actually matter — like whether students can tell their own story and make decisions they believe in?

Making sure AI tools support student thinking rather than replace it. We're studying how language models can help people analyze their own stories without overriding their judgment.

Center for Career Design. Where we run the Career Design Lab, DartWorld, and most of our programs. Seth manages programs there, which means we're working with undergrads every term.

School of Architecture. Adjunct faculty. Teaching and running career development programs with students in architecture, landscape architecture, and urban design.

Dartmouth's student-led technology lab. Our development partner for DartWorld — a new team of student engineers and designers every term, building and rebuilding the platform.