Overview
SkillBridge is a full-stack, AI-driven resume enhancement platform empowering users to upload resumes, receive intelligent feedback, store job descriptions, and automatically rewrite resumes using a RAG + RL pipeline. Users can generate LaTeX and Markdown versions, compile PDFs, and push files to GitHub via Jenkins CI.
🌐 Tech Stack
Frontend (Next.js + Tailwind CSS)
- React (with `use client` components)
- Clerk.js for authentication
- Framer Motion & Anime.js for animations
- PDF rendering with `@react-pdf/renderer`
- File handling with `react-dropzone`
Backend (FastAPI)
- REST API built with FastAPI
- Google Gemini for AI feedback and rewriting
- Weaviate Cloud for vector storage (RAG)
- LaTeX compilation and role-based filename generation
- Secure Jenkins webhook trigger for GitHub pipeline automation
Resume Processing Pipeline
- Upload Resume → PDF uploaded, parsed, stored on disk
- Parse and Embed → Text embedded & stored in Weaviate
- Feedback Generation → AI agent returns resume feedback
- Job Description Storage → Added to RAG context
- Resume Rewrite → RAG-powered Gemini agent rewrites resume
- LaTeX/Markdown Generation → Stored as `.tex` & `.md`
- LaTeX Compilation → `.tex` compiled to `.pdf`
- Jenkins Pipeline Trigger → Pushes `.tex` & `.pdf` to GitHub
⚙️ Jenkins CI/CD Setup
Pipeline
Triggered via `/api/trigger-publish`, clones GitHub repo, copies files, commits, and pushes to `main`.
Configuration
- GitHub credentials stored as `github-creds`
- Global Git config added in pipeline
- Safe directory explicitly set
- File names sanitized using custom utility
🧠 AI & Vector DB
Model: Google Gemini Pro
Storage: Weaviate Cloud (Resume → `Resume` class, JobDescription → `JobDescription` class)
RAG workflow queries nearest neighbors for tailored resumes. RL signals stored from user thumbs-up/down feedback.