Are you having a headache from working on thesis? Try new project for cyber security major.

Duck — Sat, 05 Sep 2026 04:53:00 +0000

Welcome to the newest live document for students, we are selling it with a "student price"! Who could possibly resist a great project at a cheap price?
Property:
Intruder Monitoring System

Overview

The Intruder Monitoring System is an advanced security application designed to protect workstation environments, monitor unauthorized access attempts, and maintain comprehensive violation logs through automated facial recognition, system-level locking, and cloud storage integration.

Key Features

Live Facial Detection and Anti-Spoofing: The client interface uses face-api.js to process webcam video streams, measure head pitch ratios to prevent static image spoofing, and verify face descriptors. Also, this system have live detection.
System-Level Security Control: The backend server manages workstation security by updating the Windows Winlogon shell registry, terminating explorer.exe, and deploying a locked browser kiosk interface.
Cloud Storage and Database Integration: Captured intrusion snapshots are automatically uploaded to Cloudinary, while violation records, timestamps, and reasons are persistently stored in Firebase Firestore.
Administrative Management Dashboard: Authorized users can authenticate via facial verification or a master password to access a centralized dashboard, view violation logs grouped by date, and execute individual or bulk deletion tasks. This dashboard use vercel middleware to secure the database

Property

The system uses Windows encryption devices to protect your data from being read by intruders.
The system disables the blue screen to prevent attackers from reaching safe mode, which could otherwise help them disable the system.
Furthermore, in the Learn.md file, we instruct you on how to set up a BIOS passcode. Because the BIOS is the root system used to start your PC or laptop, attackers can exploit it to plug in a bootable virus USB or reinstall Windows.
Launcher.exe, the shell file that replaces your user interface, is written in the C language, providing massive speed for the system to lock all keys and block access to the command prompt. After you unlock your device, this shell file is terminated and explorer.exe is started.
Finally, this system runs locally; an internet connection is not necessary to unlock your device. Internet access is only required for uploading the intruder's image to the cloud. If there is no internet connection, the system saves it in local storage until the computer reconnects to the network.

Data Flow

pc-lock-system folder
index.html (Intruder's image) -> server.js -> Cloudinary -> server.js (save link to database) -> firestore
Monitoring_system folder (Watch pictures)
index.html (request) -> intruder.js -> firestore -> intruder.js -> index.html
Buy now at: Tech Store: https://www.getly.store/product/pc-lock-system-for-2d-camera-with-live-detection

How I Scaled a Full-Stack E-Commerce Platform with Groq AI, Vercel Middleware, and 3D Design

Duck — Fri, 23 Jan 2026 14:26:26 +0000

Building a project is easy. Scaling it to be secure, intelligent, and visually immersive is where the real challenge begins. As a 14-year-old developer, I wanted to push the boundaries of what a "simple" e-commerce site could be. Here is how I transformed Sofas King into a high-performance solution.

The Speed of Groq AI
In this update, I moved beyond standard models and integrated Groq AI. By leveraging Groq's ultra-fast inference speed, the platform now performs real-time content moderation for comments and username changes with near-zero latency. It handles higher rate limits, ensuring the AI is always snappy and reliable for every user.

The Middleware Shield
Security cannot be an afterthought. I implemented Vercel Edge Middleware to act as the ultimate gatekeeper for the backend.

Custom Rate Limiting: I built a sophisticated API throttling system. Whether it is posting a comment or making an AI request, the system prevents bot abuse and spam.

Backend Security: Critical functions like identity updates are now protected at the server level, moving away from vulnerable client-side-only logic.

3D Visual Experience
Static images are the past. I’ve integrated 3D design elements into the UI to create a premium, tactile feel. This allows users to visualize furniture in a space that feels more "real," bridging the gap between digital browsing and physical shopping.

Architecture at a Glance
Modular Logic: 70+ ES6 files keeping the codebase clean and maintainable.

Real-time Database: Powered by Firebase Firestore with event-driven listeners for instant updates.

Performance: Edge-side execution via Vercel for the fastest possible global response times.

Open Source & Links
I believe in building in public. You can check out the full source code and the live demo below:

GitHub Source code: https://github.com/Duc-tech-hub/Sofas-king/tree/main

Live Demo: https://public-gamma-brown.vercel.app/

Video Walkthrough: https://youtu.be/KDoMIuZI0E4

Learning Journey
The biggest lesson I learned? Trust the architecture. Moving to an event-driven model and securing the "edges" of the app changed how I think about full-stack development.

webdev #javascript #ai #groq #programming #opensource #ecommerce

DEV Community: Duck

Are you having a headache from working on thesis? Try new project for cyber security major.

How I Scaled a Full-Stack E-Commerce Platform with Groq AI, Vercel Middleware, and 3D Design

webdev #javascript #ai #groq #programming #opensource #ecommerce