LLM for Reverse Engineering
Interpretation and description of unknown data.
Training on Synthetic Data
No bad examples and fully traceable.
Compatible with Templates
Integrate into any security or development tool.
Dealing with Unknown Data in Cybersecurity, from exploits to incidents
Engineers, analysis and researchers encounter unknown data in many contexts in cybersecurity. Whether it’s searching for vulnerabilities, understanding incidents and their causes, collecting and interpreting traces or using proprietary or undocumented interfaces.
This always leads to large amounts of manual work: reverse engineering of data structures. It’s mostly guessing, experimenting and just trying. This work is highly complex, time-consuming and error-prone. But there is a better way: we’re building AutoMetal, and LLM that automatically analyzes and describes provided binary data.
AutoMetal, an LLM to analyze and describe unknown binary data
Instead of complicated manual reverse engineering of data, just use the data as a prompt and let AutoMetal generate potential and plausible descriptions. This will require less samples of data to collect and cut down drastically on endless experimenting, guessing and scripting.
It will generate code directly using templates to integrate into any security analysis tool or integrated development environment, letting you parse the previously unknown data immediately.
Interested?
Sign up to be the first to know when AutoMetal is released!
This is us!
An experienced team specializing in the analysis and improvement of binary data file formats.
@Hague Security Delta (HSD)
Wilhelmina van Pruisenweg 104
2595 AN, The Hague, The Netherlands
contact@infix.ai
Send us your thoughts!
infix-tech
Check out our Github profile.
infixtechnologies
Stay in the loop – join us on LinkedIn.