Papers, bylines, press, and talks - mostly on AI agents, security operations, and where the two collide.
arXiv preprint · April 2026 · Chona, Kozlov, Kumar (lead author)
The only public benchmark for AI in security operations. It measures how well LLM agents perform the core SOC analyst task of threat hunting - and shows frontier models topping out around 45% on real security tasks. It anchors how I think about building agents that improve past that ceiling.
arXiv preprint · July 2024 · co-author
An open-source framework for building agents that decompose and execute tasks with structured, memory-infused reasoning. Cited by independent academic work and widely adopted - 460+ GitHub stars and 60K+ package downloads.
Unite.AI · April 2026
A ground-level look at how AI actually changes the anatomy of an attack - from context-aware reconnaissance to polymorphic malware - and what defense looks like when the attacker is machine-speed.
Security Boulevard · February 2026
Alert fatigue is treated as a tuning problem. It is really a capacity problem - and the fix is not fewer alerts but a different operating model.
SecurityWeek · January 2026
Quoted in SecurityWeek's flagship annual expert series on where offensive security is headed.
Help Net Security · December 2025 · video
A short expert segment on reframing threat intelligence so it drives decisions instead of adding to the noise.