Recently I've been working on:

• MCP Server for IDA Pro to allow LLMs to interact with IDA directly: ida-pro-mcp
• A beginner-friendly build system based on CMake and TOML: cmkr
• Interactive web dashboard to monitor and analyze your coding agent API costs: agent-cost-dashboard
• Python bindings for the LLVM-C API using nanobind: llvm-nanobind
• Boot loader and emulator for real mode written in C/Python: x86-real-mode-bootloader

See the pinned repositories below for other interesting things I've been up to.

Before boarding a flight I tasked my local model with a static analysis job on a whim. When I landed an hour later, the result was waiting: "I finished the requested static analysis of the sample with IDA. It's DarkSide ransomware and I fully marked up the database." Turns out the cost was $0.08 🤯

I've spent the past few months doing almost nothing but working with AI coding agents. No job, just me and LLMs building things together. What follows is everything I wish someone had told me when I started: the mindset shifts, the hard-won lessons, and the techniques that actually work for me.

In recent years the interest in obfuscation has increased, mainly because people want to protect their intellectual property. Unfortunately, most of what’s been written is focused on the theoretical aspects. In this article, we will discuss the practical engineering challenges of developing a low-footprint virtual machine interpreter. The VM is easily embeddable, built on open-source technology and has various hardening features that were achieved with minimal effort.

I am available for trainings:

• Binary Lifting with Remill (on-site)
• Payload Obfuscation for Red Teams (on-site)
• Build Your Own Coding Agent (virtual)

You can reach out for more information.

If you like or rely on the work I do, please consider sponsoring me. I do open source in my free time and it would be really great if I could do more of it.