Claudiu-Vlad Ursache (@ursachec) header
Claudiu-Vlad Ursache (@ursachec) profile picture

security research engineer @RevEng_AI

Berlin, Germany
Joined March 2010
Photo by Claudiu-Vlad Ursache (@ursachec)Photo by Claudiu-Vlad Ursache (@ursachec)Photo by Claudiu-Vlad Ursache (@ursachec)Photo by Claudiu-Vlad Ursache (@ursachec)Photo by Claudiu-Vlad Ursache (@ursachec)
Claudiu-Vlad Ursache retweeted
I'm pleased to announce a new release of the #idalib #Rust bindings for @HexRaysSA IDA Pro! This release provides new APIs and bug fixes. Thanks to new and existing contributors @mcbbugu, Amaan Qureshi, and @0xdea for their work on this release. github.com/idalib-rs/idalib.…
2
19
90
5,074
Claudiu-Vlad Ursache retweeted
today, we @RevEng_AI released the next generation of our decompiler: Ventris the new model comes with improved struct layout recovery, better PE support, and more. on decbench, we beat all other available decompilers in structure recovery, and achieve #2 rank overall
2
19
3
107
18,355
Claudiu-Vlad Ursache retweeted
I'm pleased to announce a new release of the #idalib #Rust bindings for @HexRaysSA IDA Pro! This release adds compatibility with latest SDK (v9.4), and a couple of bug fixes. Thanks to @0xdea and @yeggorv for their contributions. github.com/idalib-rs/idalib.…
21
96
8,086
Claudiu-Vlad Ursache retweeted
We are approaching perfect binary decompilation, and, crazier still, LLMs may soon be the best decompilers on the planet. I'd like to introduce DecBench, an evaluation site to determine how close we are to completing the field of perfect decompilation. Links and more in 🧵
21
134
15
928
106,123
Claudiu-Vlad Ursache retweeted
Really exciting to see the latest research from the RevEng.AI team tackling one of cybersecurity's biggest challenges: detecting malicious code at scale. As attackers continue to evolve, signature-based approaches alone are no longer enough. This research highlights how AI can identify malicious behaviour in compiled binaries by understanding the underlying semantics of machine code, enabling the detection of previously unseen malware variants without relying on constant rule updates. reveng.ai/blog/detecting-mal…
3
6
895
Claudiu-Vlad Ursache retweeted
Gradient descent does not work. I will die on this hill.
238
330
66
5,208
339,765
Claudiu-Vlad Ursache retweeted
We're pleased to announce a new release of our #Rust bindings for @HexRaysSA IDA Pro! This release adds compatibility with latest SDK, and introduces a Rust-native interface for developing plugins. github.com/idalib-rs/idalib
2
17
63
4,889
Claudiu-Vlad Ursache retweeted
What sets BRS apart is its flexibility and transparency. It is configured through product-, organization-, or ecosystem-specific risk profiles and is built to incorporate a wide range of existing metrics.
4
5
502
Claudiu-Vlad Ursache retweeted
Relying on a single metric like CVSS or EPSS can miss critical product or organizational context. BRS brings consistency to comparing different risks, such as a known high-severity vulnerability with a PoC versus a potential zero-day.
1
3
4
502
Claudiu-Vlad Ursache retweeted
Why create another metric? Traditional scoring systems are often rigid, opaque, and narrow in scope. They may not reflect product-specific requirements, differences across ecosystems such as firmware and cloud containers, or the realities of an evolving threat landscape.
1
3
5
393
Claudiu-Vlad Ursache retweeted
BRS unifies diverse security signals into a single normalized risk score, making it easier to compare otherwise disparate finding types side by side. By combining existing models and metrics, it delivers a more consistent way to assess risk across environments.
1
3
4
370
Claudiu-Vlad Ursache retweeted
source code for XTRIDE, our type recovery lib in rust, is now online! incl. docs, data pipelines and instructions to reproduce our results, and a standalone poc github.com/pr0me/xtride
we will present our work on practical binary type recovery at ACM CODASPY '26! prior work rarely focused on applicability: too slow, no meaningful filtering mechanisms, unsound struct recovery we address these gaps and present a system for headless type recovery at scale
12
1
25
4,710
Claudiu-Vlad Ursache retweeted
we will present our work on practical binary type recovery at ACM CODASPY '26! prior work rarely focused on applicability: too slow, no meaningful filtering mechanisms, unsound struct recovery we address these gaps and present a system for headless type recovery at scale
2
24
3
79
29,891
Claudiu-Vlad Ursache retweeted
As part of our vulhunt.re release, we've also dropped few rules! First up is a rule to detect a #LogoFAIL variant (CVE-2023-40238) in a UEFI DXE driver for BMP decoding... Test case(s): github.com/vulhunt-re/tests/… Rule: github.com/vulhunt-re/rules/…
1
7
2
15
6,123
Claudiu-Vlad Ursache retweeted
Replying to @francesco_ev
@francesco_ev is presenting VulHunt at @REverseConf today! Alongside the talk, we're also releasing the framework and tooling as open-source. vulhunt.re
1
11
21
1,467
Claudiu-Vlad Ursache retweeted
In the last blog of our VulHunt series, we explore what happens when VulHunt meets LLMs. We show how LLM-guided workflows can automate vulnerability triaging and hunting, reducing the manual effort required for binary analysis. binarly.io/blog/agentic-vuln…
1
15
22
2,488
Claudiu-Vlad Ursache retweeted
In today's instalment of our VulHunt series, we walk through VulHunt’s capabilities in detail. We show what VulHunt can detect, explain how it performs analysis, and how it represents findings for maximum actionability. 👉 binarly.io/blog/vulhunt-in-d…
1
15
16
1,946