Blog
Architecture, security, and the future of enterprise AI agents.
The Standards and Frameworks Behind TAOS
Every item in the word cloud is a standard or framework TAOS implements, enforces, or emits as a first-class artifact — from AES-256-GCM and Zero Trust to GDPR, SOC 2, and Rego. Here's what each one does and why it's there.
Patrick Farry
The Valiant Knights of Taos: Defenders of Trust — Episode 1
In a realm plagued by chaos, the Order of Taos remains the last bastion. Sir Valerius rallies his knights to protect legitimate exchange against theft, subversive agents, and the swell of moral decay. The TAOS way endures.
Patrick Farry
Structure First: What Building TAOS Taught Me About AI Velocity
I wrote about skeleton architecture on InfoQ. Then I built 120,000 lines in six weeks with Claude. Here's what I got right, what I missed, and why Architecture Decision Records turned out to be as important as base classes.
Patrick Farry
Why Prompt-Based Security Fails: Our NIST Submission on Governing AI Agents
In March 2026 we submitted a formal response to NIST's RFI on AI Agent Security. Our central argument: prompt-based security is structurally insufficient, and enterprises need infrastructure-level governance that treats the LLM as an untrusted actor.
Patrick Farry · Akon Dey, Ph.D. · Abhishake Gajja