AI¶
Articles on artificial intelligence — trust, reliability, development practices, and the gap between what AI can produce and what we can verify.
A0022 — AI Sycophancy: The Yes-Man Problem
Published: ~2026-04-02
There is nothing I have hated more in my career than yes-men engineers. After decades of building sycophancy-free engineering teams, the most powerful tool I have ever worked with has sycophancy baked into its core.
Background research:
R0040 — RLHF Alternatives and Sycophancy Link
Active · 2 queries
R0041 — Enterprise Sycophancy Products and Deployment
Active · 3 queries
R0042 — Private AI Motivations and Sycophancy
Active · 3 queries
R0043 — Cross-Domain Sycophancy Vocabulary
Active · 3 queries
R0044 — Expanded Vocabulary: Regulatory, Harms, Bridges
Active · 4 queries
R0048 — Corporate AI Training Gaps
Active · 3 queries
Claim verification:
R0055 — Claim Verification (iteration 1)
28 claims · 93% pass rate · 3 corrections + 5 softenings
R0056 — Claim Verification (iteration 2)
28 claims · 93% pass rate · 2 corrections
R0057 — Claim Verification (final)
33 claims · 100% pass rate
A0019 — Prompt Engineering Is Not. Engineering, That Is.
Published: ~2026-03-26
I went looking for engineering. I found creative writing advice with a technical veneer.
Background research:
R0020 — Prompt Engineering Gaps
Active · 4 queries
R0021 — Engineering Definitions and Standards
Active · 8 queries
R0023 — Counterproductive Advice and Prompt Lifecycle
Active · 4 queries
R0024 — Sycophancy, Addiction, and Vendor Incentives
Active · 4 queries
R0027 — Multilingual Prompt Engineering Challenges
Active · 3 queries
Claim verification:
R0028 — Claim Verification
Active · 33 claims
A0005 — The Trust Chasm: When AI Delivers False Confidence
Published: ~2026-03-10
We are building an increasing dependence on systems we are simultaneously losing the ability to verify.
Claim verification:
R0105 — Claim Verification
Active · 30 claims
A0007 — AI Made Everyone Faster. But Faster Is Not Necessarily Better.
Published: ~2026-03-11
You just gave everyone in the building a chainsaw. Including the people who were already a danger with scissors.
Claim verification:
R0007 — Claim Verification
Active · 15 claims
A0006 — Building on Fire: Engineering Trust in AI Systems
Published: ~2026-03-12
This is the part of the trilogy where the heroes are supposed to win.
Claim verification:
R0006 — Claim Verification
Active
A0010 — Seven Rounds With a Confident Liar
Published: ~2026-03-11
I was mid-sentence in a conversation with Claude Code when shift+enter stopped inserting newlines and started submitting my prompt instead.
Claim verification:
R0010 — Claim Verification
Active
A0008 — The Agent That Didn't Know Itself
Published: ~2026-03-03
I needed my AI coding agent to behave consistently. Same commit message format every time. Same branching workflow. Same validation steps.
Claim verification:
R0008 — Claim Verification
Active
A0011 — The Brilliant Developer, Unreliable Operator
Published: ~2026-03-03
I gave the AI agent one simple instruction: when something fails, stop what you're doing and come talk to me.
Claim verification:
R0011 — Claim Verification
Active
A0016 — Speed Has Numbers. Trust Has Error Bars.
Published: ~2026-03-03
One developer, working alone on a greenfield project, building a polyglot API ecosystem across eight repositories over forty-eight days.
Claim verification:
R0016 — Claim Verification
Active
A0009 — Building pymqrest: A Case Study in AI-Assisted Development
Published: ~2026-02-10
I built pymqrest almost entirely with AI — Claude and Codex — and I wrote virtually none of the code by hand.
Claim verification:
R0009 — Claim Verification
Active
A0004 — Global Infrastructure Diversity Trends
Published: ~2026-02-03 · Tags: technology
There is a reality here that infrastructure leaders cannot ignore.
A0002 — Why Overthinking Is a Requirement
Published: ~2026-02-01 · Tags: history
"Overthink by default" is not a personality quirk. It is a requirement.
A0001 — From Skeptic to Convert
Published: ~2026-01-29 · Tags: technology
In three months, I went from a hardcore anti-AI skeptic to someone who has essentially decided I won't be wasting my time writing code anymore.