CareerCompass AI

Main UI
Showing details of AFCAT examination of India
Chat and Ai context

Inspiration

Career decisions shape years of a person's life, yet the information needed to make those decisions is often scattered across government websites, PDFs, recruitment portals, exam notifications, forums, and confusing online advice.

As students we understand this dilemma. One may know that they want a stable public-sector career but have no idea where to start. Which exams are they eligible for? Which opportunities match their education? Which certifications would improve their chances? Which path is realistic for their age?

We saw a simple but frustrating reality: finding opportunities is hard, but understanding which opportunities are right for you is even harder.

CareerCompass AI was built to bridge that gap.

Rather than creating another chatbot that simply answers questions, we wanted to build a system that helps users explore possibilities, understand eligibility, and make informed career decisions with AI acting as a guide and not a replacement for human judgment.

What it does

CareerCompass AI is an intelligent career guidance platform focused on government and public-sector opportunities.

Users can explore examinations from different countries, review eligibility requirements, understand educational qualifications, age criteria, nationality requirements, attempt limits, category relaxations, physical standards, and other important information that often requires navigating multiple sources manually.

Beyond search and discovery, CareerCompass AI acts as a personalized career advisor.

By understanding a user's education, interests, goals, and background, the system provides tailored guidance on:

Government examinations
Public-sector careers
Skill development
Certifications
Educational pathways
Long-term career planning

Instead of providing generic answers, the platform adapts recommendations based on the user's context, helping transform overwhelming information into actionable next steps.

How we built it

We built CareerCompass AI as a Next.js application with a TypeScript frontend and backend API routes.

The frontend contains the exam explorer, eligibility dashboard, profile context panel, chat assistant, and responsible AI disclaimer. Firebase Firestore is used to cache country and exam data so the app does not need to regenerate the same information repeatedly.

The backend uses AI to help retrieve and structure exam information, extract profile context from user messages, and generate career guidance. The chat system passes conversation history, profile context, and country context to the AI assistant so responses can be personalized.

We also added a human-in-the-loop design: users choose the country, inspect exams, ask follow-up questions, and are reminded to verify final eligibility and deadlines on official government or exam authority websites.

Responsible AI

CareerCompass AI is a decision-support tool, not a final authority.

On every launch, the app shows a disclaimer explaining that AI responses and cached exam information may be incomplete, outdated, or incorrect. Users are instructed to verify eligibility rules, application deadlines, and official notices through government or exam authority websites.

The assistant is limited to career, government jobs, government exams, education pathways, certifications, and professional development. Users can clear their chat and profile context.

Challenges we ran into

One challenge was keeping the AI grounded in the correct country context. For example, if a user explored exams in one country but asked a career question about another, the assistant could become confused. We improved this by prioritizing the user’s latest message and extracted nationality when setting the assistant’s country context.

Another challenge was designing the app so it felt useful immediately, instead of just being another chatbot. We solved this by pairing chat guidance with a structured exam dashboard and profile context panel.

We also had to balance helpful personalization with privacy. The app extracts only lightweight profile context and includes controls to clear saved context.

Accomplishments that we're proud of

Building a complete end-to-end AI career guidance platform within the hackathon timeline, from concept to deployment.
Creating an experience that goes beyond a traditional chatbot by combining structured examination discovery, eligibility exploration, and personalized AI guidance in a single workflow.
Designing a human-in-the-loop system where AI assists users without replacing their judgment, encouraging verification through official sources and responsible decision-making.
Developing context-aware career recommendations that adapt to a user's education, interests, goals, and background instead of producing generic responses.
Successfully combining structured information and generative AI to make complex government and public-sector career pathways easier to understand.
Implementing responsible AI guardrails, transparency notices, and user controls so personalization remains useful without becoming intrusive.
Building a scalable architecture that separates user experience, AI reasoning, and data storage, making the platform easier to expand in the future.
Transforming fragmented career information into a guided exploration experience that helps users move from uncertainty toward actionable next steps.

What we learned

We learned that the biggest value of AI in public-service navigation is not just answering questions. It is helping users connect scattered information into a practical next step.

We also learned that responsible AI has to be visible in the product, not hidden in documentation. That is why CareerCompass includes an upfront disclaimer, official-source reminders, and human verification as part of the user flow.