Essays
The Journal253 essays
Long-form writing on AI, philosophy, psychology, and systems thinking.
-
Data Lineage as Ethical Infrastructure
End-to-end data lineage tracking revealed that data used in customer-facing decisions had undocumented transformations nobody could explain. Lineage is a moral obligation when data affects people.
-
Sartre’s Bad Faith and The Algorithm Decided
Saying 'the algorithm decided' is bad faith. Every algorithm encodes human choices, and delegation does not eliminate moral responsibility.
-
The AI Ethics Career Path Does Not Exist Yet
AI ethics job postings demand technical skills the role never uses, while rarely mentioning process design, the skill that matters most.
-
The Stoic Programmer: Marcus Aurelius on Technical Leadership
Marcus Aurelius governed a vast empire during plague and war. His Stoic principles map to engineering leadership: control what you can and serve your team.
-
The FinOps Problem in AI Agent Systems
A plan-and-execute pattern that routes most agent tasks to cheaper models cuts inference costs steadily while keeping task accuracy intact.
-
AI Code and the 322% Privilege Escalation Problem
Apiiro found AI-generated code carries more privilege escalation vulnerabilities, so AI coding demands more engineering discipline, not less. AI coding tools demand more engineering discipline, not less.
-
The Problem of Other Minds and Large Language Models
The problem of other minds holds that we cannot prove consciousness exists in others. LLMs intensify this by producing behavioral signatures of understanding. Large language models sharpen the old problem of other minds. The question of machine consciousness may be permanently unanswerable.
-
Process Debt Is More Expensive Than Technical Debt
Process debt consumed more capacity than technical debt in the engineering organizations I studied, and it compounds faster because it affects everyone.
-
The Rise of the Vertical AI Stack
Vertically integrated AI applications shipped features faster than horizontally decomposed systems. The vertical AI stack represents a return to integration that microservices proponents should study.
-
The Social Media Ethics Problem Is an Attention Architecture Problem
Moderation catches only a fraction of harmful content. The ethics problem lives in the algorithms that decide what gets amplified. The larger ethical problem is the attention architecture that amplifies content optimized for engagement over wellbeing.