Inspiration Career confusion is personal for us. We have seen students choose degrees because they were told they were “safe,” only to realize too late that they hated them. Friends spent years and money on paths they never used. Others skipped college without a clear alternative because no one showed them real options. Career guidance today is broken. Counselors are overloaded, online quizzes are shallow, and college is treated as the default answer even when it isn’t the right one. By the time people understand themselves, they have already lost time, money, or motivation. We wanted to build something better — something people would actually engage with.
The Idea (Innovation) Career Path AI helps users discover careers, degree options, and no-college pathways through gamified, adaptive questioning, then generates a personalized roadmap to reach those paths. Instead of static quizzes, the system runs a diagnostic conversation. Short, Duolingo-style questions adapt based on user responses and gradually surface interests, strengths, and constraints. What makes this different is that all paths are treated equally — college, bootcamps, certifications, apprenticeships, and self-taught routes. We don’t just suggest careers; we show users how to realistically get there.
How It Works (Technical Implementation) The frontend is built with React, using a clean, dashboard-style interface that feels conversational rather than form-like. On the backend, Node.js handles user flow and sessions, while FastAPI (Python) powers the AI layer. We use Gemini AI to drive adaptive questioning, interpret user signals, and generate personalized career roadmaps. Each answer contributes to a profile covering interests, strengths, tolerances, and constraints. To make the experience more human and accessible, we integrated ElevenLabs for optional voice output. All results appear instantly in the dashboard — no PDFs, no long reports, just clear next steps.
Challenges & Limitations Ensuring AI reliability was a major challenge. Early versions sometimes looped or produced unrealistic suggestions, so we added guardrails to keep outputs consistent and grounded. Design was equally hard. Early versions worked but felt robotic, so we rewrote prompts and microcopy to make the experience feel supportive and human. We are college freshers and were not physically in the same place, so all collaboration happened remotely while balancing classes. Time and resources were also a constraint: we couldn’t afford paid API keys for some services, especially the pro versions, which limited the number of requests and slowed development. Because of the hackathon time limits, we also reduced career coverage, skipped session persistence, and limited regional data. With more time and resources, we would expand careers, improve personalization, and add saved progress.
Impact & Utility Career Path AI is built for students choosing what’s next, college students questioning their major, career switchers exploring new paths, and non-traditional learners who need affordable options. This matters most in underserved and fast-changing job markets, where wrong decisions are costly and guidance is limited.
Scaling & Future Plans The system is modular and scalable. We plan to expand career data, support more regions and languages, and partner with schools, NGOs, and employers. Long-term, we see Career Path AI as a tool that helps align education with real-world opportunities without forcing everyone into the same path.
Demo & Presentation The demo starts immediately with interactive questions — no signup wall. As answers change, the system adapts in real time and generates a personalized dashboard with careers and roadmaps. It feels like a conversation, not a quiz.
Closing Career Path AI is a working prototype built to solve a real problem. It helps people understand their options earlier, explore all valid paths, and move forward with clarity instead of guesswork. We are proud of what we built and excited about where it can go ne
