I ran the same reverse engineering “skills” across OpenAI Codex and Claude Code in a FLARE-VM workflow to see which one handles RE skills better. This post breaks down the results for IOC extraction and unpacking, where the tools differed, what that means for defensible artifacts, and what I’d change to make the comparison truly apples-to-apples.
Using AI skills as reusable playbooks for reverse engineering.
My journey to improve my reverse engineering skills is just beginning.
A build 20 years in the making.
Using the AI IDE from AWS to deploy infrastructure quickly and securely.
Setting up a self sustained network for my rental property IoT devices.
Exploring how AI can speed up reverse engineering workflows. I walk through my environment setup, testing steps, and insights on integrating LLMs into malware analysis.
Growing Without Losing Your Edge
Making the tool more flexible and usable.
Let’s use Ghidra to examine some sample malware.
OK, let’s do some basic testing with DeepSeek.
I’ve been diving deep lately into reverse engineering and malware analysis, a field that I’ve only had a limited exposure to in my day-to-day activities. Part of the learning is focused on writing simple assembly code. I’ve typically focused on high-level languages such as Python and Golang, so the syntax of assembly code is a fun learning curve. My starting point was learning the different ELF…
“You really need this person to mentor you.”
“How do you scale security policies and controls for developers?”
"If I have to look it up twice, I'm writing it down."
This will be a space where I post about things I learn (or have learned), projects I’m working on, research I’ve performed, or just overall tech knowledge. I’m looking forward to using this space often for these updates, and this has been a long time coming.