- AI agent securityI assess AI agent deployments before they get access to your files, email, or systems: what the agent can reach, how it can be tricked, and what you'll see when it goes wrong.You get: an assessment of what the agent can do, where it could go wrong, and the limits to put in place.
- Threat huntingI produce threat intelligence, run hypothesis-driven hunts, and grade every finding by confidence and source quality.You get: a threat assessment report with a confidence rating on each finding.
- HardeningI write security benchmarks and review your configurations against recognized baselines.You get: a prioritized list of configuration findings, each tied to the benchmark.
Each engagement produces the writing it needs: assessments, benchmarks, detection content, and analytic reporting. I take every engagement myself.
Show four more tools, including two for planning AI hardware
I take part in open, royalty-free standards work on how AI agents share memory, and I maintain my own independent implementation. I do this on my own behalf.
I spent five years on Army Cyber Protection Teams, then worked as a senior threat hunter supporting DHS. I hold the CISSP and five GIAC certifications covering incident response, forensics, malware analysis, threat intelligence, and open-source intelligence. I also serve on the GIAC Advisory Board.