Skill2Job

Inspiration

Many students learn through scattered courses, YouTube videos, and tutorials but still struggle to understand what skills they are missing for their target job. We wanted to build a platform that connects a student's current knowledge with the skills actually required for a career and provides a personalized path to become job-ready.

What it does

Skill2Job assesses a student's current knowledge through assessments and quizzes, builds a personalized skill profile, identifies knowledge gaps, and recommends what to learn next. It also uses learning resources and relevant content to support each recommendation. As the student progresses, the platform continuously updates their skill profile and readiness level, helping them understand where they are now, what they need to improve, and how close they are to their target career.

How we built it

We built Skill2Job as a full-stack learning and career-readiness platform.

Frontend: Interactive dashboard for assessments, learning paths, quizzes, progress, and skill visualization. Backend: Handles user data, assessments, recommendations, and application logic. Database: Stores user profiles, assessment results, skills, progress, and learning information. RAG: Retrieves relevant learning content based on the student's identified knowledge gaps. LLM: Helps generate personalized explanations, questions, and learning guidance. ML: Uses student performance and skill-related data to estimate career readiness. Adaptive learning: Quiz difficulty and recommendations can change according to the student's performance.

Challenges we ran into

One of our biggest challenges was deciding how to accurately identify a student's actual skill level instead of relying only on self-assessment. We also had to figure out how to combine assessments, learning resources, retrieval, personalized recommendations, and readiness prediction into one consistent workflow. Another challenge was making the recommendations useful rather than simply generating a large list of topics. The system needed to focus on the student's actual knowledge gaps and career goal.

Accomplishments that we're proud of

We are proud of creating a complete workflow that connects assessment → skill-gap identification → personalized learning → adaptive practice → progress tracking → career readiness. Instead of treating learning as a fixed course sequence, Skill2Job attempts to adapt the learning journey according to each student's current abilities and progress. We also successfully combined retrieval, language models, machine learning, and structured student data into a single practical use case.

What we learned

We learned that personalization is not just about generating different content for every user. Good personalization requires reliable assessment data, meaningful skill representation, relevant resources, and continuous feedback. We also learned that different technologies should have clear roles. Retrieval is useful for finding relevant information, language models are useful for interaction and generation, while machine learning can help analyze structured performance data. Most importantly, we learned that solving a real problem requires designing the complete user journey rather than focusing only on individual technologies.

What's next for Skill2Job

Our next goal is to make Skill2Job more accurate, practical, and connected to real career opportunities.

We plan to:

Improve skill assessment and readiness prediction using more real-world data. Add more career paths and skill frameworks. Improve the quality and freshness of recommended learning resources. Add project recommendations based on skill gaps. Track progress toward specific job requirements. Provide personalized interview and placement preparation. Explore integration with real internship and job opportunities.

Ultimately, we want Skill2Job to become a continuous career-readiness companion that helps students move from learning skills to actually demonstrating and applying them.

Built With

• llm
• machine-learning
• natural-language-processing
• ollama
• python
• qwen
• rag
• scikit-learn
• sql
• supabase
• technology