PathWise AI — Career & Advising Action Layer
A student-first platform that turns academic and career uncertainty into maps, personalized recommendations, roadmaps, and coach-ready next steps.
- ↗ Built a career exploration and advising product around maps, roadmaps, and personalized next steps
- ↗ Positioned the product as an action layer for students, advisors, career centers, and departments
- ↗ Opened early pilot/collaboration conversations with student-facing programs and ecosystem partners
- ↗ Turned the KeyVoid / Advisor AI vision into a clearer institutional product direction
Overview
PathWise AI began from a simple pattern I kept seeing around students: ambition was not the issue. Direction was.
Students were surrounded by links, advisors, degree pages, role descriptions, clubs, certificates, and career platforms, but the real question remained unresolved: what should I actually do next, and why does that step fit me?
PathWise is my answer to that problem. It is a student-first action layer for academic and career navigation. Instead of only showing information, it helps users explore majors and careers, visualize pathways, generate personalized roadmaps, and leave with a coach-ready summary that can support advising and career conversations.
The problem
Most student-success tools create more surfaces, not more clarity. Students are expected to piece together:
- degree requirements
- career outcomes
- skill gaps
- certificates
- internships
- advising notes
- long-term plans
- weekly next steps
That creates a gap between information access and action readiness. PathWise is designed to close that gap.
Users and stakeholders
The product is built for multiple layers:
- students who need clarity around majors, careers, and opportunities
- advisors who need better-prepared students
- career centers that want actionable coaching summaries
- departments that want pathway visibility and student engagement
- programs that need customizable career exploration experiences
The core design principle is that the student experience should be simple, while the institutional layer should be configurable.
My role
I led the product direction, positioning, early workflow design, demo strategy, and full-stack implementation direction. My work included shaping the product narrative, scoping the main flows, coordinating pilot conversations, testing the experience, and turning the broader KeyVoid / Advisor AI vision into a more focused PathWise product.
Product decisions
The most important decision was to frame PathWise as an action layer, not another dashboard.
That meant prioritizing:
- Exploration — helping users discover possible directions.
- Interactive maps — making pathways visual instead of buried in text.
- Personalized roadmaps — turning interest into semester, career, or weekly plans.
- Coach-ready summaries — helping students bring better context into advising or career conversations.
Technical and operational approach
The early platform was built with a pragmatic full-stack architecture: Python/Flask, Jinja templates, data-backed recommendation flows, Cytoscape.js for interactive map experiences, and deployment on cloud infrastructure. The system was designed around fast iteration, demo readiness, and the ability to customize for different programs.
The goal was not to overbuild the platform before validation. The goal was to get a credible version in front of users, advisors, departments, and ecosystem partners quickly enough to learn from real conversations.
Impact
PathWise moved from concept to live demos, early usage, and pilot conversations. I presented the direction publicly during AZ Tech Week’s Tech Talent Summit 5.0 + Startup Pavilion and continued refining the product around the needs of students, advisors, and institutional partners.
The clearest traction signal was that people understood the pain quickly: students do not just need more career information. They need a path they can act on.
What I learned
The biggest learning was that an AI product only becomes valuable when it changes the user’s next action. A recommender that produces a list is not enough. A useful product has to connect exploration, reasoning, planning, and handoff.
That shaped how I now think about AI products: the model is only one part of the system. The real product is the workflow around the model.
PM / APM interview story
Situation: Students had access to career and advising information, but many still lacked clarity on what to do next.
Task: Build a student-facing product that could turn scattered information into personalized direction and useful advising artifacts.
Action: I scoped PathWise around exploration, interactive maps, personalized recommendations, roadmaps, and coach-ready summaries. I worked on the product architecture, demo flow, and pilot positioning, then put the product in front of real stakeholders through startup and university ecosystem channels.
Result: PathWise became a clearer product direction with live demos, early pilot conversations, and a stronger narrative around becoming an action layer for student success.