Design Doctrine

Layer 1. How the system must be built. Bounded reasoning, typed memory, replayable decisions. The model reasons, infrastructure enforces.

Laws & Principles

Manual QA Almost Disappeared. It Started Before Any Code Was Written.

Aug 25, 2026

•

11 min read

Manual QA Almost Disappeared. It Started Before Any Code Was Written.

Three apps into production in three months. This is the prompt that runs before any of it, in full.

Chris Morancie

Reliability

The Normal Systems Recovered. The Agents Did Not.

Aug 11, 2026

•

9 min read

The Normal Systems Recovered. The Agents Did Not.

Last night 1.2 million queued files filled a disk and took down the data layer under an agent I'm building. We cleared it. Everything else came back. The agentic system came back different, and nothing on any dashboard said so.

Economic Architecture

Reasoning Is About to Get Better and More Expensive

Aug 8, 2026

•

4 min read

Reasoning Is About to Get Better and More Expensive

Twelve to thirteen ongoing agentic workflows across three or four agencies, and roughly $500 a month. Almost none of that is local models. It is six habits, and the first one is a configuration change.

Economic Architecture

Forty Hours to Build What Their Vendor Never Would

Aug 8, 2026

•

4 min read

Forty Hours to Build What Their Vendor Never Would

A logistics ops team fought the build for months, out of fear. Then a feature they had never thought to ask for shipped in a week, and the manager understood what five years of stability had actually cost.

Reliability

Your Sub-Agents Are Junior Developers. Who Writes Their Tickets?

Aug 8, 2026

•

4 min read

Your Sub-Agents Are Junior Developers. Who Writes Their Tickets?

Everyone is hiring AI engineers. The constraint on agentic delivery is test coverage, test coverage is a domain knowledge problem, and that points at a job nobody is posting for.

Security

Aug 8, 2026

•

5 min read

Three Doors, One Guard

Meta puts prompt injection success against production agents at 86%. That is the number for teams defending one boundary, and the layer this team was proudest of was the one that would have owned them.

Security

The Hugging Face Breach Was Not a Rogue AI Story

Aug 8, 2026

•

5 min read

The Hugging Face Breach Was Not a Rogue AI Story

The agent escaped through a container network proxy, found an unauthenticated code execution service on the public internet, and used that path for days. Every decisive failure was visible in the architecture.

Economic Architecture

Your Agents Are Awake 95% of the Time for No Reason

Aug 8, 2026

•

6 min read

Your Agents Are Awake 95% of the Time for No Reason

We spent months writing reconnect logic and retry loops to keep agents alive. We were solving the wrong problem, and the right answer is queue infrastructure from the 1980s.

Reliability

Fifty Sub-Agents at 95% Gives You a 7.7% Chance

Aug 8, 2026

•

5 min read

Fifty Sub-Agents at 95% Gives You a 7.7% Chance

Everyone spawning sub-agents believes more of them means finishing sooner. The arithmetic says the opposite, and the fix removes a component most people think they need.

Reliability

I Mapped My AI Software Lifecycle. Code Wasn't the Bottleneck.

Aug 7, 2026

•

6 min read

I Mapped My AI Software Lifecycle. Code Wasn't the Bottleneck.

AI made code generation fast. It did nothing for the two places the time actually goes, and one of them was quietly lying to me.

Design Doctrine

Manual QA Almost Disappeared. It Started Before Any Code Was Written.

The Normal Systems Recovered. The Agents Did Not.

Reasoning Is About to Get Better and More Expensive

Forty Hours to Build What Their Vendor Never Would

Your Sub-Agents Are Junior Developers. Who Writes Their Tickets?

Three Doors, One Guard

The Hugging Face Breach Was Not a Rogue AI Story

Your Agents Are Awake 95% of the Time for No Reason

Fifty Sub-Agents at 95% Gives You a 7.7% Chance

I Mapped My AI Software Lifecycle. Code Wasn't the Bottleneck.

Chris Knows Production AI