Skip to content

Media

Talks and commentary on AI security, from someone who tests the systems.

Kevin O'Connor is available for technical discussions about agent permissions, connected tools, software security and laboratory systems. Send the topic, the audience, the format and a proposed date, and we'll tell you whether it's a fit and whether he's free.

Speaker information

Background and topics for your program.

  • Short and extended biographies
  • Suggested technical topics
  • Links to published work and professional background
  • Inquiries for podcasts, panels and private briefings

Short bio

Kevin O'Connor is founder and Principal Engineer of TKOResearch. He assesses AI systems, conducts applied security research, and builds software for technical and scientific work, including TAREOps.

Full bio

Kevin O’Connor is founder and Principal Engineer of TKOResearch. His work covers AI-agent permissions, tool access and software security, alongside development of TAREOps, the company’s AI-native Laboratory Information Management System. He previously worked on CrowdStrike’s research team and led threat research at Adlumin and N-able. He writes about technical methods and system behavior for engineering teams and advises leadership and counsel on security questions.

01

Topic areas

  • AI-agent authority and control planes
  • Prompt injection, MCP/tool abuse, RAG leakage, and CI/CD agent risk
  • AgentBoundary and AI-agent robustness testing
  • Technical review for boards, counsel, and investors
  • Lab records, access control and TAREOps

Next step

Send the topic, the format and a date.

Request availability