just another security-related blog.
Hey, I’m Igris — a security engineer who likes taking things apart to understand how they break.
This blog is my field notebook: deep dives into Linux and kernel internals, hardware security, reverse engineering and malware development, and lately the messy, fascinating frontier where AI meets offensive and defensive security.
Expect hands-on write-ups, paper breakdowns, and the occasional rabbit hole — written to make hard problems a little more approachable (and to remind me what I learned along the way). Whether you’re a fellow practitioner, a curious beginner, or just here to watch something get rooted, pull up a chair.
Let’s explore, break, and build.
Recent blogs,
-
The End of Online Anonymity? How LLMs Are Cracking the Code of Practical Obscurity
For decades, internet users have relied on a comforting shield known as “practical obscurity.” The idea is simple: while you could theoretically be identified by your zip code,...
-
Can a 7B Model Beat GPT-o3 at Finding Bugs? Meet VulnLLM-R
In the world of cybersecurity, finding vulnerabilities is like finding a needle in a haystack. Traditionally, we relied on static analysis tools (like CodeQL) which are fast but...
-
Securing the Agentic Future: A Deep Dive into AI-Agent Protocol Threats
The evolution of Artificial Intelligence has been nothing short of remarkable. We have moved from the rigidity of Symbolic AI and Expert Systems to the pattern-matching capabilities of...
-
Prompt Injection is Dead. Long Live Promptware: The 7-Stage Kill Chain
For the past few years, the cybersecurity community has comforted itself with a familiar analogy: Prompt Injection is just the LLM version of SQL Injection.
-
Finding Backdoors in LLMs Using Their Own Memory
Large Language Models (LLMs) are becoming the backbone of modern software. But what if the model you just downloaded has a secret agenda?