🎓 SkillBridge

The Bridge Between Academia and Industry — Rebuilt from the Ground Up

A centralized, blockchain-verified portal for intelligent opportunity discovery, skill-gap analysis, and end-to-end academic project lifecycle management.

🚀 Live Demo · 📹 Video Walkthrough · 📄 Docs

---

📌 Problem Statement

The bridge between academia and industry is currently fragmented. Internship opportunities are scattered across siloed job boards, and academic projects often lack real-world relevance because students cannot easily find industry mentors or datasets. There is a significant skills gap mismatch, and recruiters struggle to verify the authentic contribution of students in group-based academic work.

AcadLedger is our answer to this challenge — a unified, AI-powered, blockchain-verified platform that closes the academia-industry gap at every stage of a student's career journey.

---

🎯 Objective

Create a centralized, blockchain-verified portal for:

• 🔍 Intelligent, skill-based opportunity discovery
• 📊 Automated skill-gap analysis with AI
• 🗂️ End-to-end project and application lifecycle management
• 🤝 Transparent, verifiable student-recruiter collaboration

---

✅ Expected Outcomes (How We Deliver Them)

Expected Outcome	AcadLedger Implementation
Skill-based, hyper-personalized opportunity matching using NLP-driven resume-to-JD parsing	AI Resume Analyzer — GPT-4 class LLM parses resumes against JDs, produces ATS scores, keyword gap analysis, and section-by-section feedback
Simplified, transparent application tracking and milestone management	Recruiter ATS Kanban Board — drag-and-drop pipeline with status: pending → reviewing → shortlisted → hired/rejected
Strengthened industry-academia collaboration through co-sponsored projects	Recruiter Postings Portal — recruiters publish skill-tagged opportunities; students apply directly within the platform
Higher student placement rates and verifiable digital project portfolios	Blockchain-Verified Profiles — SHA-256 hashed academic portfolios with Git-scraper integration for authentic contribution tracking

---

🌟 Key Features

🤖 AI-Powered Resume Analyzer

• Upload resume PDF → select a Job Description → get instant AI analysis
• Returns: ATS Score (0–100), matched/missing keywords, skill gap breakdown (technical, tools, soft), section-level feedback, improved bullet points, and a shortlist verdict
• Multi-model fallback chain (Gemini 2.5 Flash → Llama 4 → Gemini 2.0 → DeepSeek → Llama 3.3 → Qwen3) via OpenRouter — zero downtime even under rate limits
• PDF text extraction with pdf-parse; truncation safety for large resumes

🕷️ Git Scraper — Authentic Contribution Tracking

• Scrapes public GitHub repositories linked to academic projects
• Extracts commit history, file-level contributions, and PR/issue activity
• Maps contributions per-student in group projects — solving the "free rider" problem
• Results are hashed and stored as verifiable blockchain records

🔗 Blockchain-Verified Portfolios (SHA-256)

• Every academic milestone, test score, project contribution, and certification is hashed using SHA-256
• Hash is stored on-chain, making credentials tamper-evident and independently verifiable
• Recruiters can one-click verify any student's portfolio authenticity
• Eliminates resume fraud and ghostwritten project submissions

📋 Recruiter ATS Kanban Board

• Full drag-and-drop Kanban (pending → applied → reviewing → shortlisted → rejected → hired) built with @dnd-kit
• Optimistic UI updates with server-side sync
• Applicant detail drawer with resume, cover letter, and profile
• Bulk status management, search, and filter

📢 Opportunity Posting & Discovery

• Recruiters create structured postings: skills required, stipend, duration, location/remote, deadline, max applications
• Optional skill gate: attach a proctored test that candidates must pass before applying
• Students browse and filter opportunities matched to their verified skill profile

🧪 Proctored Skill Assessment Engine

• Recruiters build dynamic MCQ tests (single correct, multiple correct, true/false)
• Features: shuffle questions/options, strict mode (tab-switch detection), configurable time limits, negative marking
• Server-side auto-grading via PostgreSQL stored procedures (grade_attempt_v2)
• Session resume — if a student loses connection, they pick up exactly where they left off

📈 Analytics Dashboards

• Recruiter dashboard: live/upcoming/past tests, total attempts, draft stats, placement rates
• Candidate dashboard: completed tests, available opportunities, ATS score history, skill progress

👤 Dual-Role Architecture

• candidate — Students with academic profiles, skill tests, resume uploads, applications
• recruiter — Companies/institutions with test creation, postings, ATS pipeline, analytics

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                        Frontend                             │
│   Next.js 15 (App Router) + React 19 + Tailwind CSS v4     │
│   ├── /auth         — Sign in / Sign up                    │
│   ├── /~/home       — Role-aware dashboard                 │
│   ├── /~/jobs       — Opportunity discovery (students)     │
│   ├── /~/postings   — ATS management (recruiters)         │
│   ├── /~/tests      — Proctored assessment engine          │
│   ├── /~/resume     — Resume builder                       │
│   ├── /~/resume-analyzer — AI resume ↔ JD parser          │
│   ├── /~/analytics  — Placement analytics                  │
│   └── /~/candidates — Talent pool (recruiters)             │
└─────────────────────────────────────────────────────────────┘
                         │
                         ▼
┌─────────────────────────────────────────────────────────────┐
│                       Backend                               │
│   Supabase (PostgreSQL 15 + Auth + Storage + RLS)          │
│   ├── Row-Level Security on every table                    │
│   ├── Stored Procedures (grade_attempt, save_test_v2…)     │
│   ├── Triggers (auto-profile creation, session sync)       │
│   └── Edge Functions for webhook integrations              │
└─────────────────────────────────────────────────────────────┘
                         │
                         ▼
┌─────────────────────────────────────────────────────────────┐
│                    External Services                        │
│   ├── OpenRouter AI  — Multi-LLM resume analysis           │
│   ├── Git Scraper API — GitHub contribution extraction     │
│   └── SHA-256 Hasher — Blockchain credential verification  │
└─────────────────────────────────────────────────────────────┘

Database Schema Highlights

Table	Purpose
profiles	Unified user identity (candidate / recruiter)
candidate_profiles	Academic details, skills, institution linkage
recruiter_profiles	Company info, branding
opportunities	Job/internship postings with skill requirements
applications	Application lifecycle with status tracking
tests	Proctored assessments linked to opportunities
questions / options	MCQ bank with tagging
test_attempts	Secure attempt tracking with expiry & resume
attempt_answers	Per-question answer records for grading
blockchain_records	SHA-256 hashed credential store
git_contributions	Per-student GitHub activity from scraper

---

🛠️ Tech Stack

Layer	Technology
Framework	Next.js 15 (App Router, Server Actions)
Language	TypeScript 5
UI	Tailwind CSS v4, Radix UI, shadcn/ui
Animations	Motion (Framer Motion v12), dnd-kit
Database	Supabase (PostgreSQL 15)
Auth	Supabase Auth (JWT + OAuth)
AI/LLM	OpenRouter (Gemini, Llama, DeepSeek, Qwen)
PDF Processing	pdf-parse (server-side)
Blockchain	SHA-256 (Node.js crypto)
Git Scraper	Python Flask API + PyGitHub
Export	jsPDF + jspdf-autotable
Deployment	Vercel (Frontend) + Supabase Cloud

---

🚀 Getting Started

Prerequisites

• Node.js 20+
• A Supabase project
• An OpenRouter API key (free tier works)

Installation

# Clone the repository
git clone https://github.com/pushkar2510/AcadLedger.git
cd AcadLedger

# Install dependencies
npm install

# Configure environment variables
cp .env.local.example .env.local

Environment Variables

# Supabase
NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_anon_key

# AI (OpenRouter)
OPENROUTER_API_KEY=your_openrouter_key

# Git Scraper Backend
NEXT_PUBLIC_GIT_SCRAPER_URL=http://localhost:5000

Database Setup

# Apply the full schema to your Supabase project
# Go to Supabase Studio > SQL Editor and run:
schema_recruiter.sql   # Full production schema

Run Locally

npm run dev
# App runs at http://localhost:3000

Running the Git Scraper

cd git-scraper
pip install -r requirements.txt
python app.py
# Scraper API runs at http://localhost:5000

---

🔐 Security & Trust Architecture

Row-Level Security (RLS)

Every database table is protected by PostgreSQL RLS policies — users can only access their own data. Recruiters cannot see other recruiters' candidates; candidates cannot see other candidates' applications.

Proctoring

• Tab-switch detection with server-recorded tab_switch_count
• Strict mode locks the browser viewport
• Server-side expiry: if a student's timer runs out, the attempt is auto-submitted and graded server-side via a stored procedure — no client-side manipulation possible

Blockchain Verification (SHA-256)

Student completes milestone
        ↓
Platform serializes: { student_id, milestone_type, result, timestamp }
        ↓
SHA-256 hash generated on server
        ↓
Hash stored in `blockchain_records` table
        ↓
Recruiter verifies: recomputes hash from public data → compare → ✅ or ❌

This makes every score, certificate, and project contribution independently verifiable without trusting AcadLedger's servers.

---

📂 Project Structure

AcadLedger/
├── app/
│   ├── (dashboard)/~/
│   │   ├── home/            # Role-aware landing dashboard
│   │   ├── jobs/            # Opportunity discovery (students)
│   │   ├── postings/        # Recruiter posting management + ATS Kanban
│   │   ├── tests/           # Test creation & attempts
│   │   ├── resume/          # Resume builder
│   │   ├── resume-analyzer/ # AI resume ↔ JD analyzer
│   │   ├── analytics/       # Placement analytics
│   │   ├── candidates/      # Recruiter talent pool
│   │   └── settings/        # Profile & preferences
│   ├── auth/                # Sign in / Sign up
│   └── page.tsx             # Marketing landing page
├── components/
│   ├── app-sidebar.tsx      # Navigation sidebar
│   ├── feature-section.tsx  # Landing page features
│   └── ui/                  # Design system components
├── lib/
│   └── supabase/            # Server & client helpers
├── schema_recruiter.sql     # Full production database schema
└── schema.sql               # Original schema dump

---

🤝 Team

Built at [Hackathon Name] — [Date]

Name	Role
[Team Lead]	Full-Stack & AI Integration
[Member 2]	Blockchain & Git Scraper
[Member 3]	UI/UX & Frontend
[Member 4]	Database & Backend

---

📊 Impact Metrics (Target)

• 🎓 Students — Reduce time-to-first-internship by 40% via personalized matching
• 🏢 Recruiters — Cut screening time by 60% with AI-ranked, skill-verified candidates
• 🔒 Trust — 100% verifiable credentials via SHA-256 blockchain hashing
• 🌐 Opportunities — Real-time aggregation from GitHub activity + job postings

---

📜 License

This project is licensed under the MIT License — see LICENSE for details.

---

Built with ❤️ to close the gap between students and their dreams.

AcadLedger — Where academic achievement meets industry opportunity.