Developer • Creator • Problem Solver
Subarna Basnet is a Full-Stack Developer, Software Engineer, Game Server Architect, and Security Engineer based in Kathmandu, Nepal. Welcome to Subarna OS, an interactive personal operating system portfolio featuring software engineering projects, free developer tools, and Nepali Unicode & Patro widgets.
Highly skilled Full-Stack Developer and Game Server Architect with strong expertise in FiveM/QBCore development, backend automation, and security engineering. I specialize in building systems that feel alive—dynamic UIs, optimized server logic, and realistic gameplay mechanics. Experienced in designing scalable systems including roleplay servers, monitoring dashboards, and security scanners.
A Python-based security tool that detects obfuscated Lua patterns, backdoors, and malicious loaders in FiveM resources. detailed logging via Discord webhooks. Built with Python, Regex, Discord API, Security.
Web-based system monitoring dashboard for Linux VPS. Visualizes network traffic, manages services, and includes a secure file manager. Built with Python, Flask, Linux, MySQL.
An environmental project using Google Earth Engine (GEE) to detect and map unmanaged waste disposal sites. Built with Google Earth Engine, Python, GIS.
A machine learning application utilizing Convolutional Neural Networks (CNN) to predict stock price trends based on historical data. Built with Python, CNN, TensorFlow, Pandas.
Deepfake video detection system designed to analyze and identify manipulated media content. Built with Python, Deep Learning, OpenCV.
Built a custom game server with Lua scripting on a VPS, featuring optimized resource management and custom gameplay loops. Built with Lua, VPS, Networking.
Subarna Basnet specializes in Lua / FiveM / QBCore, Python / Flask, JavaScript / Node.js, React.js, Linux / VPS Management, SQL / MariaDB, Security / Malware Analysis, DevOps & CI/CD, C++ / IoT. Explore 18+ browser developer tools and 15+ Nepali Unicode, Patro, EMI, Tax, Land, and NEPSE tools built into Subarna OS.
Leading Artificial Intelligence and Machine Learning initiatives. Overseeing research and development of AI-driven solutions.
Managed deployment pipelines, server infrastructure, and operational workflows to ensure high availability and security.
Designed and maintained large-scale FiveM roleplay servers using QBCore and Lua. Implemented anti-cheat logic, NPC traffic systems, and optimized OneSync entity handling.
Oversaw IT operations and production workflows. Managed team coordination and technical resource allocation.
Developed and maintained mobile applications, focusing on performance and user experience.
Roleplay servers are prime targets for malicious scripts. In this guide, I break down how to detect obfuscated Lua code and secure your resources...
Lag is the enemy of immersion. Learn how to optimize entity handling and server-sided loops to support 100+ players without FPS drops.
Creating believable NPC traffic goes beyond spawning random vehicles. I walk through designing AI-driven traffic flows, NPC pilots, air traffic at LSIA, and cleanup logic to keep s
Obfuscated Lua is often used to hide malicious intent in FiveM scripts. In this post, I explain common obfuscation patterns, encrypted loaders, and how I built a Python-based stati
Static portfolios are boring. This article explains my approach to building a modular developer portal that changes layout, theme, and interaction patterns on every refresh while s
Running a public VPS invites constant attacks. I cover practical steps to harden Ubuntu servers including firewall rules, service isolation, SSH protection, logging, and monitoring
Manual server maintenance does not scale. In this guide, I show how to automate backups, monitoring, cleanup tasks, and health checks using Python scripts scheduled via cron jobs.
Anti-cheat systems should be lightweight and smart. This post explores detecting illegal mods, memory manipulation patterns, abnormal player behavior, and how to balance security w
Visibility is key to stability. I break down how I built a Flask-based VPS monitoring dashboard that tracks network usage, running services, logs, and system health in real time.
Streaming workflows can be automated. This article explains how I designed an RTMP-based pipeline using OBS and custom backend logic to streamline video uploads and live streaming
As AI-generated content grows, so does the need for verification. A look into my research on using Convolutional Neural Networks (CNN) to detect artifacts in deepfake videos with h
Before deploying any app, you need a secure foundation. Learn the essentials of UFW firewalls, SSH key authentication, and Fail2Ban configurations to protect your Ubuntu servers.
Why settle for a static site? I break down how I built my "living" developer portal that changes layouts on refresh and fetches real-time GitHub stats using a custom API.
From chat apps to live notifications, WebSockets are essential. This tutorial walks through building a scalable chat server that handles private messaging and rooms efficiently.
How we used GEE and Python to detect unmanaged waste disposal sites. A case study on leveraging satellite imagery for environmental impact analysis.
Stop writing spaghetti code. Learn how to structure your FiveM resources using modular patterns, events, and exports to make your server code maintainable and scalable.
Financial markets are chaotic, but patterns exist. I share my experience building a stock prediction dashboard using Long Short-Term Memory networks and Flask.
Reflecting on my role as AI & ML Head at ITIC. We discuss the growing tech landscape in Nepal, the adoption of AI in local industries, and the road ahead for developers.
Deep Fake Detection — Project by Subarna Basnet — About me in Subarna OS by Subarna Basnet (Full-Stack Developer | Game Server Architect | Security Engineer).
Deepfake video detection system designed to analyze and identify manipulated media content.
Keywords: Python, Deep Learning, OpenCV