Path Finder: Adaptive Career Guidance System

Inspiration

Millions of students and job seekers face career confusion, often picking paths based on short-term market hype or peer pressure rather than their true personal strengths and intrinsic interests. Most traditional career guidance tools provide one-time static quizzes with vague, generic advice. We wanted to build an intelligent, interactive mentor—an AI-driven platform that dynamically adapts its evaluation, identifies genuine strengths, flags skill gaps, and lays down a structured, actionable step-by-step roadmap from beginner to industry-ready professional.

What it does

Adaptive Career Assessment:** Conducts a context-aware 6-step interactive assessment where follow-up questions adapt dynamically based on prior responses.
NLP Interest Classification:** Categorizes student aptitudes across domains such as Technology, Business, Creative, Education, Sports, and Government.
Strength Profile & Persona:** Automatically generates a comprehensive breakdown of user strengths, suggested career title, and personal motivation style.
Structured Roadmaps:** Creates customized 3-tier milestone paths (Beginner → Intermediate → Advanced) covering foundation skills, practical hands-on projects, and industry readiness.
Skill Gap Analysis:** Measures readiness percentages, identifies missing proficiencies, and offers targeted recommendations.
Secure Persistent Dashboard:** Keeps user progress, assessment results, and activity history synced via persistent authentication and database storage.

How we built it

Frontend:** Built with Next.js (App Router), React, TypeScript, Tailwind CSS, and Radix UI components for an accessible, responsive, and modern user experience.
Backend & APIs:** Next.js Serverless API routes running on Node.js to power fast, scalable REST endpoints.
Database:** MongoDB Atlas with Mongoose schemas managing 1:1 and 1:N relations for Users, Assessments, Roadmaps, and Activity Logs.
AI & NLP Engine:** Rule-based decision algorithms and NLP classifiers integrated with Google Gemini LLM to analyze student responses and craft tailored guidance.
Authentication & Security:** JWT in HTTP-only cookies, password hashing with bcrypt, and Google OAuth 2.0.

Challenges we faced

Dynamic Adaptive Logic:** Ensuring assessment questions adapt intelligently without repeating or drifting away from the core aptitude domain.
Data Normalization:** Structuring roadmaps for vastly different fields (e.g., Tech vs. Creative vs. Sports) into a unified, measurable milestone format.
State & Session Persistence:** Designing a smooth user flow that maintains assessment state and user authentication seamlessly across page reloads.

Accomplishments that we're proud of

Developing a full-stack, end-to-end working career engine that generates tangible, structured roadmaps rather than generic text outputs.
Seamlessly integrating adaptive question logic with persistent database storage.
Building a sleek, intuitive dashboard that makes career planning actionable and motivating for students.

What we learned

How to design effective prompt and classification pipelines combining deterministic NLP rules with LLM intelligence.
Optimizing Next.js App Router performance with serverless API architecture and MongoDB connection pooling.
The importance of user-centric UX design when presenting complex career guidance data.

What's next for AI Career Guidance

-> AI Resume Builder & Reviewer:** Parsing resumes to auto-fill skill gaps directly against the generated roadmaps. -> Real-Time Internship & Job Matching:** Connecting users with opportunities aligned with their readiness scores. -> AI Career Voice Coach:** Adding conversational voice-based interview practice and mentorship.