Skip to content
View Ishani-Shree's full-sized avatar
4 followers · 1 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report Ishani-Shree

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Ishani-Shree/README.md

Typing SVG

👩‍💻 whoami

ishani@shree:~$ cat profile.json
{
  "name"      : "Ishani Shree",
  "pronouns"  : "she/her",
  "education" : [
    "BTech CSE @ SRMIST '27 ",
    "BS Data Science @ IIT Madras '28"
  ],
  "roles"     : [
    "AI/ML Developer",
    "Full-Stack Engineer",
    "Offensive Security Enthusiast"
  ],
  "leadership": [
    "Team Head · Team Envision, Aaruush",
    "Committee Head · Highlights, Aaruush",
    "Super Coordinator · Paradox, IIT Madras"
  ],
  "location"  : "Chennai, India"
}
 
ishani@shree:~$ cat current_focus.txt

🤖 AI/ML Systems
        → Transfer Learning · EfficientNet-B2
        → Scikit-learn · PyTorch · TensorFlow
        → Production-grade ML Inference

🌐 Full-Stack Development
        → React 18 · TypeScript · Tailwind CSS
        → FastAPI · Node.js · PHP
        → MySQL · PostgreSQL · MongoDB

🛡️  Offensive Security
        → XSS · SQLi · CSRF · SSRF · CMDi
        → Penetration Testing · CTF Practice
        → Web Security Research

🛠️ Technology Arsenal

Languages

AI / ML

Frontend

Backend & Databases

Security & Tools

DevOps & Cloud

🚀 Featured Projects

🤖 Underwater Microplastic Detection

AI/ML · Computer Vision · Production Deployment

End-to-end AI classification system achieving 97.3% validation accuracy across 5 polymer classes. EfficientNet-B2 with ImageNet transfer learning, AdamW + CosineAnnealingLR — deployed as a real-time Flask inference dashboard.

Python PyTorch EfficientNet Flask

🎯 97.3% validation accuracy
🔬 5 polymer classes classified
⚡ Real-time inference dashboard
🌊 Environmental monitoring ready

Repo

🏢 PlaceTrack — Placement Management System

Full-Stack · AI-Integrated · Role-Based System

Full-stack recruitment lifecycle platform with 4 role-based portals — mirrors an end-to-end H2R workflow. Ships an AI-powered Resume Analyser with ATS scoring and a live analytics dashboard.

React TypeScript PHP MySQL

👥 4 role-based portals
🤖 AI Resume Analyser + ATS scoring
📊 Live placement analytics dashboard
🔐 Session auth + role-scoped access

Repo

🩺 Diabetes Risk Analyser

Clinical AI · FastAPI · Full-Stack

Full-stack AI decision-support tool processing 8 clinical parameters to deliver instant Low/Moderate/High risk classification with confidence scores and contributing-factor breakdowns.

Python FastAPI React Scikit

⚕️  8 clinical parameters processed
🎯 Low / Moderate / High classification
📈 Confidence scores + factor breakdown
🔬 Recharts data visualizations

Repo

🛡️ Vulnerable Web Application

Offensive Security · Penetration Testing · Educational

Deliberately vulnerable web app covering 7 OWASP attack vectors — built for penetration testing practice and security awareness training. Full login system with exploitable endpoints.

NodeJS Express MySQL

XSS SQLi CSRF SSRF CMDi

⚠️  Educational use only
🎯 7 OWASP attack vectors
🔓 Exploitable endpoints per vuln

Repo

👁️ Face Recognition Attendance System

Computer Vision · On-Device ML · Python

End-to-end CV pipeline — captures 100 face samples per student, trains a local LBPH model on-device, and deploys real-time inference that auto-marks attendance with no cloud dependency.

Python OpenCV LBPH Tkinter

📸 100 face samples per student
🔒 85% confidence threshold
☁️  Zero cloud dependency
✅ Auto attendance marking

Repo

🛒 ShopZone — E-Commerce Platform

Full-Stack · JWT Auth · PostgreSQL

Full-stack e-commerce platform with JWT authentication, 92 products across 14 categories, payment simulation (card + COD), order management, and real-time INR currency conversion.

React NodeJS PostgreSQL JWT

🛍️  92 products · 14 categories
💳 Card + COD payment simulation
💱 Real-time INR currency conversion
📦 Full order management system

Repo

📊 GitHub Stats

Ishani's GitHub Stats Ishani's Streak Stats
Ishani's Top Languages

🌐 Let's Connect

LinkedIn GitHub Email

💭 "The best way to predict the future is to build it."


Visitor Count


Thanks for visiting! Feel free to explore, fork, and collaborate. 🚀

Pinned Loading

  1. PlaceTrack PlaceTrack Public

    A full-stack college placement management system with role-based portals for Admin, Students, Recruiters, and Placement Officers. Built with React 18 + TypeScript + Tailwind CSS frontend and PHP 8.…

    JavaScript

  2. ShopZone ShopZone Public

    A full-stack e-commerce platform with JWT auth, 92 products across 14 categories, payment simulation (card + COD), order management, and real-time INR-based currency conversion. Built with React 18…

    JavaScript

  3. Underwater_Microplastic_Detection Underwater_Microplastic_Detection Public

    AI-powered underwater microplastic detection and classification using EfficientNet-B2. Classifies 5 polymer types (alkyd, bead, cellophane, degraded, fiber) with 97.3% accuracy and a real-time Flas…

    Python

  4. Airline_Reservation_System Airline_Reservation_System Public

    Desktop airline reservation system built with Java Swing and MySQL — real-time booking, live flight tracking via OpenSky Network, simulated payment, and boarding pass generation.

    Java 1

  5. Vulnerable-Web-Application Vulnerable-Web-Application Public

    A deliberately vulnerable web application for security training and CTF practice. Built with Node.js, Express & MySQL. Covers OS Command Injection, SQL Injection, XSS (GET/POST), CSRF (GET/POST), a…

    HTML

  6. Face_reco Face_reco Public

    Desktop attendance system using real-time face recognition. Register students via webcam, train a local LBPH model, and mark attendance automatically — no cloud, no manual entry. Built with Python,…

    Python