Placivio

AI-powered campus placement platform connecting students, TPO and companies through Gemini — giving every student a personalized roadmap to get placed.

Comment

Inspiration

I am a 6th semester Computer Science & Business Systems student at GGITS, Jabalpur. Every single day I watched my batchmates struggle with the same three questions: "What should I learn next?", "Which companies are coming to campus?", and "Am I even ready for placements?"

The answer was always the same — nobody knew. Students learned randomly, TPOs managed drives on WhatsApp groups and Excel sheets, and companies arrived on campus with zero idea of which students were actually ready for them.

I built Placivio to fix all three problems at once.

What It Does

Placivio is a 3-sided AI-powered campus placement platform built on Google Gemini and MongoDB Atlas:

For Students:

  • Enter your branch, semester, CGPA and current skills
  • Gemini AI generates a personalized week-by-week learning roadmap with free resources
  • Real-time skill gap analysis showing exactly what you're missing for your target companies
  • Placement readiness score (0-100) that updates as you progress
  • Campus drive alerts the moment your TPO posts a new company visit
  • AI chat coach that knows your full profile and answers questions like "Am I ready for TCS?"
  • Weekly check-in system where Gemini reshuffles your roadmap based on your actual progress
  • Achievement badges to keep you motivated

For TPO (Training & Placement Officers):

  • Post company drives in 2 minutes — students get personalized notifications instantly
  • See every student ranked by match percentage for each specific drive
  • One-click AI shortlisting — Gemini ranks top candidates with reasoning for each selection
  • Batch analytics dashboard showing readiness distribution, skill heatmaps, at-risk students
  • AI Batch Report — Gemini generates a full paragraph analysis of your entire batch health

The Agent Loop: Student applies to drive → Gemini reprioritizes their roadmap for that company → Student improves skills → Match percentage increases → TPO sees updated ranking → Company gets best candidates.

This is not a chatbot. This is a persistent AI agent that takes actions, tracks progress over time, and connects three different users through one intelligent system.

How We Built It

  • AI Brain: Google Gemini 1.5 Flash API for roadmap generation, skill gap analysis, AI chat, candidate shortlisting and batch reporting
  • Database: MongoDB Atlas (M0 free tier) with 6 collections: students, tpos, drives, applications, notifications, achievements
  • MongoDB MCP Integration: Used MongoDB's Model Context Protocol server to enable the Gemini agent to directly query and update student data, making it a true agentic system rather than a simple API
  • Frontend: React + Tailwind CSS deployed on Vercel
  • Backend: Node.js + Express with JWT authentication
  • Charts: Recharts for placement readiness visualization

Challenges We Faced

  1. Making it truly agentic: The hardest part was making Gemini not just answer questions but actually take actions — reshuffling roadmaps, updating scores, sending notifications. This required careful prompt engineering and MongoDB MCP integration.

  2. Real match percentage calculation: Computing dynamic skill match between a student's current skills and a company's requirements in real time required careful schema design in MongoDB.

  3. Three-sided data model: Building one system that serves students, TPOs and companies with completely different needs from a single MongoDB database required significant architectural thinking.

  4. Free deployment: Built and deployed the entire platform at zero cost using MongoDB Atlas free tier, Gemini free API, and Vercel — proving this can scale without infrastructure investment.

What We Learned

  • How to build truly agentic AI systems that take actions rather than just answering questions
  • MongoDB Atlas schema design for multi-user platforms
  • How Gemini's context window enables personalized AI coaching at scale

  • The power of MongoDB MCP for giving AI agents direct database access

    What's Next for Placivio

  • Integration with LinkedIn for automatic skill verification

  • WhatsApp notifications for drive alerts

  • Multi-college support so TPOs from different campuses can onboard

  • Resume builder powered by Gemini

  • Interview preparation module with mock questions generated for specific companies

Try It Yourself

Live Demo: https://placivio.vercel.app

Demo Credentials:

🎓 Student Accounts (VIT Vellore): | Role | Email | Password | |------|-------|----------| | CSE High Readiness | aarav@student.edu | password123 | | CSE Mid Readiness | diya@student.edu | password123 | | ECE Mid Readiness | kabir@student.edu | password123 | | IT Needs Attention | ananya@student.edu | password123 |

🏫 TPO Account: | College | Email | Password | |---------|-------|----------| | VIT Vellore | tpo@vit.edu | password123 |

Start as TPO → Post a drive → Login as student → See the alert → Apply → Watch TPO dashboard update in real time.

Built With

Share this project:

Updates