ai

19 posts in this category.

Practicing What I Preach: What I Added (and Skipped) When Auditing My Own Blogging Agent

Dynamic Guardrails for Agentic AI: Why Static Rules Break Down When Agents Can Delegate

The Silent Failure Mode in Multi-Pass LLM Pipelines: When Your Model Runs Out of Tokens and Says Nothing

Delegation Is the Real Identity Problem in Agentic AI

Building an Automated Model Evaluation Pipeline: What Worked, What Didn't

Evaluations: The control plane for AI governance

I spent a Saturday debugging my AI blog agent and ended up rebuilding how it thinks

Agent Observability: The Missing Layer in Agentic AI Platforms

From Periodic Reviews to Continuous Governance

The IETF is now working on agent authentication. Here is what that means.

Weekend engineering: smarter AI pipeline alarms and upgrading to Claude Sonnet 4.6 with extended thinking

Governing Autonomous Agents Is a Platform Problem

Agents Are Not Software

Teaching the Blog Agent to See: Conceptual Diagrams and Visual Thinking

The Conversation After: Agentic AI Needs a Platform Mindset

Upgrading the Blog Agent: Sonnet 4.6, Web Search, and Bug Fixes

I Built an AI Agent That Writes, Revises, and Publishes to My Blog

Why Agentic AI Needs a Platform Mindset, Not Just Better Prompts

Agentic AI in Financial Services