Building Secure Systems. Breaking Insecure Ones.
AI Security Researcher | Red Team Specialist | SOC Analyst
NARESH RAJJ S
AI Security Researcher
The Cybersecurity
Mind Behind the Screen
Passionate about finding vulnerabilities before attackers do. Specializing in AI/LLM security research, Red Team operations, and building robust SOC detection systems. Every system has a weakness — my mission is to find it and fix it.
The Journey So Far
From writing first lines of code to breaking into systems and securing AI — every step has been a building block.
Programming
Started with Python and C. Built foundations in algorithms, data structures, and problem-solving.
Web Development
Full-stack development with React, Node.js, Flask. Built web applications from scratch.
Cybersecurity
Dove into offensive security. Mastered Burp Suite, Nmap, Metasploit. Started hunting vulnerabilities.
CTF Competitions
Competed on TryHackMe, HackTheBox, PicoCTF. Developed CTF challenges. Placed 8th at Root Access 2k26.
AI Security Research
Researching LLM vulnerabilities, prompt injection, AI red teaming. Building secure AI applications and agents.
Challenge Development
Creating CTF challenges, security tools, and contributing to the offensive security community.
Skills & Expertise
Programming
Web Development
Cybersecurity
Offensive Tools
AI & LLM Security
Platforms
Case Studies
Security tools, AI applications, and offensive research — built to break and built to protect.
Capture The Flag
Challenge archives from TryHackMe, HackTheBox, PicoCTF, and custom CTF development.
AI Security Research
Investigating vulnerabilities in AI systems, LLMs, and autonomous agents.
LLM Prompt Injection
Researching attack vectors against large language models including direct and indirect prompt injection techniques. Developing defense mechanisms and input sanitization strategies for production LLM deployments.
AI Agent Exploitation
Identifying security gaps in autonomous AI agents and developing exploitation methodologies. Mapping attack surfaces across tool-use, memory, and multi-agent orchestration layers.
Model Context Protocol Security
Analyzing security implications of MCP in AI agent ecosystems and tool integrations. Auditing permission models, tool-call validation, and cross-tenant data isolation in MCP servers.
Behavioral Threat Detection
Building statistical baselining systems for detecting anomalous authentication behavior. Implementing Z-score and IQR-based anomaly scoring with real-time alert pipelines for SOC operations.
Open Source Activity
Timeline
Credentials
Cybersecurity
AI & LLM
Write-ups & Insights
CTF write-ups, security research notes, and AI security explorations.
Get In Touch
Have a project in mind or want to collaborate? Let's talk.