PedagogyBridge

Inspiration

Millions of workers possess world-class operational and hands-on experience, yet they get automatically filtered out by corporate resume scanners simply because they don't use traditional corporate buzzwords. I wanted to solve this glaring economic barrier by building a bridge between everyday physical or service-oriented labor and formal career pathways.

What it does

PedagogyBridge is an audio-first "reverse-resume" platform. Instead of forcing users to fill out tedious text forms, they can simply speak naturally in their native language or dialect to describe what they actually do day-to-day. The platform uses OpenAI's Whisper and structured LLM parsing to instantly extract their hidden competencies, generate a professional "Competency Passport," and match them with local micro-apprenticeship or career opportunities.

How I built it

I built a robust, lightweight full-stack MVP consisting of:

• Backend: Python and FastAPI handling API routes, audio uploads, and structured JSON extraction mapped directly onto official labor market competency frameworks.
• Frontend: Next.js with Tailwind CSS, featuring a minimalist, high-impact landing page and an intuitive voice-recording interface utilizing the browser's MediaRecorder API.

Challenges I ran into

Managing audio blob formats across different web browsers and ensuring that the LLM reliably outputted clean, structured JSON schemas without breaking the frontend dashboard required careful asynchronous handling and strict error fallbacks.

Accomplishments that I'm proud of

I managed to build a fully working end-to-end AI pipeline—from live voice capture in the browser to backend speech-to-text translation and dynamic skill-matching—within a tight hackathon window.

What I learned

I gained a deeper understanding of integrating real-time audio streams with server-side AI endpoints, structuring reliable JSON outputs from large language models, and designing high-impact user interfaces for accessibility.

What's next for PedagogyBridge

I plan to expand the platform by integrating live geolocation data for hyper-local community college programs, adding multilingual support for non-native English speakers, and partnering with local workforce boards to test the platform with real job seekers.

Built With

• api
• css
• fastapi
• gpt-4o
• javascript
• next.js
• openai
• python
• tailwind
• whisper