Team 87_Career Compass

Inspiration

Fresh graduates in Singapore face an overwhelming amount of career information, but struggle to convert it into clear, actionable decisions. Job descriptions are vague, skill requirements are scattered, and it’s hard to know where to start. The unemployment rate for fresh university graduates in Singapore within six months of graduation increased to approximately 12.9% in 2024, up from 10.4% in 2023.

We were inspired by the growing graduate unemployment rate and our own experiences of feeling “almost qualified, but not quite sure what to do next.” Career Compass was built to close this gap between information and action.

What it does

Career Compass is a career planning and job-matching platform that helps users:

• Understand how well they fit specific roles through a transparent job–candidate fit score
• Identify missing skills using skills gap analysis and convert them into personalised learning roadmaps
• Provides curated courses mapped to each missing skill, so users can immediately start closing the gap.
• Explore real-world job previews beyond standard job descriptions
• Track applications and interviews in one centralized place
• Get guided career advice through an AI-powered career assistant

The platform supports onboarding via structured forms or resume upload, allowing users to quickly build a comprehensive career profile.

The AI-powered career assistant has these following roles:

• Interview Prep Mode — mock interviews, STAR method coaching, and tailored feedback
• Roadmap Mode — structured learning plans based on career goals and skill gaps
• Resume Review & Career Advice — instant feedback and improvement suggestions
• Suggested Career Paths — curated tracks such as Software Engineering, Data Analytics, and Product Management
• Advanced Filtering & Personalization — filters roles, skills, industries, work arrangements, and career paths based on user preferences

How we built it

Career Compass is a full-stack web application with a clear separation between data ingestion, core logic, and user experience. The frontend was built with React and TypeScript, featuring a multi-step onboarding flow, interactive dashboards, and real-time feedback on job fit and skill readiness.

On the backend, we implemented APIs to manage user profiles, applications, and job data, with persistent storage on Microsoft Azure. Job postings from the provided CSV datasets were parsed and integrated to power job search, filtering, and role insights.

Core features such as job–candidate fit scoring and skills gap analysis were implemented using explicit matching logic that compares user skills against role-specific requirements, driving fit scores, skill checklists, and learning roadmaps. The OpenAI API was used selectively for resume summarization and career guidance, while keeping core logic transparent and deterministic.

Challenges we ran into

Time management was one of our biggest challenges. Career Compass was built under a significantly shorter timeline than our usual projects, which typically span 4–8 weeks.

With limited time, we had to prioritize core features that delivered the most value, make fast technical decisions, and scope the platform carefully to remain feasible while still meaningful.

We also faced challenges integrating and structuring job datasets from CSV files, ensuring the data could be reliably mapped to user profiles, skills, and roles within the system.

Designing a coherent user experience across onboarding, job matching, skills analysis, and applications—while working under tight deadlines—required constant iteration and trade-offs.

Accomplishments that we're proud of

• Building an end-to-end platform that goes beyond job search into career planning
• Implementing a transparent fit score and skills gap system users can clearly understand
• Designing a structured onboarding flow that captures complex career data efficiently
• Successfully integrating provided job datasets into real product features
• Creating a demo-ready prototype with real workflows, not just static pages

What we learned

We learned that career planning is not a single decision, but an ongoing process that needs structure, feedback, and adaptability.

From a technical perspective, we gained experience in building scalable frontend components, designing meaningful data models, and implementing logic-driven features that connect across multiple parts of a system.

We also learned the importance of explaining why a recommendation exists, not just presenting the result.

What's next for Team 87_Career Compass

Next, we plan to:

• Expand job and skills datasets for broader role coverage
• Improve learning roadmap personalization based on progress over time
• Add deeper analytics for application outcomes
• Refine the career assistant with better contextual memory
• Support more career paths beyond early-career roles
• Make this into a fully developed and published website, that can also come in the form of an app

Built With

• api
• azure
• csv
• express.js
• git
• github
• microsoft
• node.js
• openai
• react
• typescript
• vscode