ASCEND

Inspiration

As CS, CSE, and Data Science students, we constantly hear advice about “preparing for the future,” but rarely get concrete, personalized guidance on what to do right now. Students juggle heavy course loads, projects, internships, and recruiting timelines, yet still struggle to understand where they stand, what they’re missing, and how close they are to being career-ready.

We wanted to build something that bridges this gap. A system that translates academic progress into clear, actionable career direction.

What it does

ASCEND is an AI-powered career advisor for CS and Data Science students that transforms coursework, skills, and experiences into:

A readiness percentage for specific roles
Personalized skill gap analysis
Step-by-step academic and career roadmaps
Curated opportunities (jobs, internships, research, post-grad programs, certifications)

Instead of generic advice, ASCEND gives students clarity on where they are and what to do next.

How we built it

We focused on designing and prototyping the end-to-end product experience rather than implementing a full production backend.

We created detailed low-fidelity and high-fidelity designs in Figma to define user flows, layout, and interactions across the platform. These designs guided the structure of a React + TypeScript frontend prototype, which we iteratively refined using Claude to generate and edit components.

Our work emphasized:

Defining data structures for skills, readiness scores, and opportunities
Designing how AI outputs would be surfaced in the UI
Simulating realistic system behavior with mock data

This approach allowed us to validate product feasibility and user experience while clearly mapping how a future backend and AI pipeline would integrate.

Challenges we ran into

Translating a complex AI-driven concept into a simple, intuitive interface
Deciding what information students actually need versus what is “nice to have”
Designing screens that balance personalization with clarity
Keeping designs consistent across multiple product surfaces

Accomplishments that we're proud of

Designing a cohesive multi-page product experience
Creating realistic prototypes that demonstrate core functionality
Clearly mapping how AI would power each feature
Producing a concept that is technically feasible and user-centered

What we learned

We learned that AI is most powerful when it augments structured systems rather than replacing them. Combining traditional ML with LLMs creates solutions that are both intelligent and trustworthy. We also gained experience in product thinking, model selection, and designing for real student needs.