Project created by Ethan Del Campo, Caleb Berent, Caius Pelish, and Nurlan Mammadli

Inspiration

Navigating career choices can be overwhelming, especially in a rapidly evolving job market. We wanted to create a tool that empowers users to self-assess their strengths and preferences and receive actionable, AI-driven career recommendations — instantly. By blending technology with career development, we hoped to bridge the gap between self-awareness and opportunity.

What it does

SkillMatch AI guides users through a 30-question self-assessment that evaluates soft skills, technical abilities, work style, and learning preferences. Once completed, the platform sends the responses to an AI backend powered by Google's Gemini model, which analyzes the input and returns three tailored career suggestions with rationale — all in real time.

How we built it

Frontend: Built with React and styled using Tailwind CSS for a clean, responsive UI.

Backend: Node.js + Express handles survey submission and calls the Gemini API using the user's responses.

AI Integration: Prompts are dynamically generated and sent to the Gemini 2.0 Flash model via the Generative Language API.

Communication: The frontend and backend communicate seamlessly via RESTful API with error handling and async state management.

Challenges we ran into

Prompt engineering: Getting Gemini to follow specific formatting instructions (e.g., no asterisks or markdown) required lots of testing and tweaks.

Response consistency: Ensuring that the AI returned clean, list-free, and relevant responses every time.

UX clarity (for frontend): Making sure users are aware of unanswered questions without overwhelming them visually.

Time constraints: Balancing functionality, polish, and technical scope; whilst delegating tasks to achieve our common goal.

Accomplishments that we're proud of

Successfully integrated a generative AI model to produce intelligent, relevant career guidance.

Created a smooth, mobile-responsive survey experience with dynamic validation and visual feedback.

Built a clean, modular, full-stack project that’s scalable and deployable.

Designed an experience that actually feels helpful and insightful — not just a gimmick.

What we learned

How to prompt generative AI effectively (and how not to).

How to design a survey UI/UX that is accessible and engaging.

How to architect full-stack applications that integrate third-party AI tools.

That even small AI-powered features can make a big impact when implemented thoughtfully.

What's next for SkillMatch AI

Add support for saving progress and re-accessing results later.

Provide additional resources (e.g., courses, certifications) tied to each career suggestion.

Add support for uploading a resume or LinkedIn profile to generate richer insights.

Expand the assessment to include personality or behavioral-style questions.

Deploy to production and invite real users to try it out!

Built With

Share this project:

Updates