Overview
News
Technologies
Salaries
Products
People
Growth
Offices
Financials
Overview
Founder of DeepTrail, building trust infrastructure for AI agents. Writing about building AI agents, AI infrastructure, agent infrastructure, AI agent security, and trustworthy AI systems. Previously Motive, NIO, D2iQ (Nutanix), E8 Security (VMware), Boundary (BMC).
News
Remediation Asymmetry: When Agents Can Diagnose More Than They Can Fix
The better your agent's diagnostic capability, the more money it wastes. 97.2% of elapsed runtime occurred after the root cause was already known.
Read more
Report
Behavioral Induction: How Capabilities Shape Agent Execution
The failure mode of an autonomous agent isn't that it does nothing. It's that it does too much. Capabilities induce behaviors - more than instructions.
Read more
Report
Constraint Durability: The Missing Layer Between Policy and Trust
Context windows are not just a performance surface. For AI agents with authority, they are part of the trust boundary.
Read more
Report
Permission Envelope Compilation: Why Static Permissions Are the Hardcoded Query Plans of Agent Systems
Same task, same agent, three permission models. 34 tools visible, only 4 used. Static permissions are starting to look like hardcoded query plans.
Read more
Report