Lunoway

Inspiration

As undergraduate students entering the workforce, we observed a critical disconnect in Singapore's data-driven market. While there is an abundance of career data available, it is often fragmented and overwhelming. We noticed that fresh graduates struggle to transform this scattered information into a coherent plan. They often ask, "What specific skills am I missing for this role?" , "How do I modify my resume to apply to a specific role" or "How do I get from here to becoming a Solutions Architect in 5 years?"

We were inspired to solve this problem by creating a unified platform, Lunoway, that doesn't just list jobs, but actively helps close the skills gap. Our goal was to democratize access to career guidance, effectively acting as a digital mentor that turns confusion into a personalized, actionable roadmap for employment.

What it does

Lunoway is a comprehensive career acceleration platform designed to bridge the gap between education and employment. It functions as both a job aggregator and a personalized career coach.

Smart Job Matching & Compatibility: Users upload their resumes, and our system analyzes them against job descriptions to provide a clear compatibility score.
Gap Analysis & Roadmaps: Instead of just saying "Not Qualified," Lunoway identifies specific missing skills and generates a Personalized Upskilling Roadmap. It suggests specific courses and immediate next steps to bridge those gaps.
Career Trajectory Planning: Users can select a long-term role (e.g., "Solutions Architect in 5 years"). The platform generates a multi-step pathway, outlining the intermediate roles and skills needed to reach that milestone.
Application Tracking: A centralized dashboard allows users to track applications across specific stages: Applied, Screening, Interview, Offer, and Rejected.
Advanced Filtering: Users can filter opportunities not just by title, but by Location, Job Type (Full-time/Remote), Salary Range, and Experience Level.

How we built it

We prioritized a high-performance, scalable tech stack to ensure a seamless user experience:

Frontend: We built the interface using React + Vite. This combination gave us a blazing fast development environment and an SEO-friendly, responsive dashboard. We utilized Lovable to accelerate our UI development.
Backend & Database: We leveraged Firebase for its robust ecosystem, utilizing it to store job roles and user profiles. We used Firebase Authentication for secure user management and Firestore for our real-time database needs.
AI Engine: The core logic is powered by the OpenAI API (GPT-4o). We engineered prompts to feed the AI a job description and a user's resume. The model processes these inputs to output a JSON object containing "Missing Skills" and "Suggested Course Keywords."
- The matching logic can be conceptualized as a set difference operation where we identify the set of missing skills between the Job Description and the Applicants.
Data Visualization: To make the data digestible, we used Recharts (and explored Chart.js) to visualize the user's skill growth and application progress over time.

Challenges we ran into

Data Unification: Transforming "fragmented career data" into a structured format was difficult. We had to implement logic to retract info from JDs and fill in missing fields].
Prompt Engineering: Getting the AI to return consistent, strictly formatted JSON for the roadmap generation was a trial-and-error process.
Scope Management: We had ambitious plans, including Telegram notifications and a complex AI Chatbot. Balancing these "nice-to-haves" with the core features like centralized data and job filtering was a constant exercise in prioritization.

Accomplishments that we're proud of

Visualizing the Pipeline: Successfully implementing the dashboard that turns abstract application statuses into a clear, visual pipeline.
Seamless Integration: Getting the Frontend (Vite) to talk perfectly with the Firebase Backend and the AI API without significant latency.

What we learned

The Power of Structured Data: We learned that the quality of AI output is directly dependent on the quality of the input data structure as well .
User-Centric Design: Building the "Transparent Requirements" feature taught us that clarity is valuable. Users prefer explicit "missing skills" lists over vague "low match" warnings.
Full-Stack Synergy: We deepened our understanding of how modern frontend frameworks interact with serverless backend solutions like Firebase as well as how to use Google Authentication.

What's next for Lunoway

Behavioral Recommendations: We plan to improve the recommendation engine to learn from user behavior patterns(e.g., which jobs they save or ignore).
Full AI Chatbot: We aim to fully flesh out the AI Chatbot to handle resume tips and application guidance.
AI Resume Generator: A feature to automatically generate a personalized resume based on a "master resume" tailored to a specific job description.
Telegram Integration: Finalizing the notification system to push job alerts and application status updates directly to users via Telegram instead of through email.