Field notes.
Real opinions about what AI is getting right and wrong for operators in 2026, written by us, in our own voice, not a content calendar. Plus the occasional teardown of something Izzy built and what it taught us.
Articles
The 4 AI "Loops of Death" That Kill Budgets (And How to Stop Them)
Autonomous agents don't crash when they fail; they burn capital. Discover the 4 AI "Loops of Death" draining your engineering budget and the necessary safeguards to stop them.
The Architecture of Autonomy: Why Human-in-the-Loop Is Permanent Infrastructure
HITL isn't temporary it's essential for Level 3 Autonomy. Learn architectural patterns like Interruption Gateways and Risk-Tiered Routing to secure Agentic AI.
LLM Evaluation Guide: How to Build a Golden Set for Prompts
Public benchmarks fail for enterprise AI. Learn the engineering protocol for building, versioning, and automating Golden Sets for reliable LLM evaluation.
The AI Solvency Crisis: Fixing Evaluation Economics with Hybrid Active Learning
Stop the vibes tax. Learn how a Hybrid Active Learning router cuts AI evaluation costs by 80% while ensuring EU AI Act compliance and data reliability.
Prompt Engineering as Code: Why "Magic Strings" Kill AI Reliability
Move beyond vibe checks. Implement Prompt Engineering as Code (PEaC) to prevent regressions, control costs, and ensure compliance with the EU AI Act. A guide for AI CTOs.
Calibrated Reliance: Stop AI Hallucinations with Better UX
Seamless interfaces make users trust AI hallucinations. Learn how CTOs can design for Calibrated Reliance using risk-weighted UI friction and ensure AI safety.
RAG Hallucinations: Why Your Vector Database Is Lying to You (And How to Fix It)
You can't prompt-engineer your way out of bad retrieval. Learn how Semantic Chunking, Metadata Enrichment, and Reranking eliminate RAG hallucinations at the source.
The Top 5 Problems with PromptMetrics (And Why You Might Want to Avoid Us)
Thinking of buying PromptMetrics? Read this honest review of our top 5 limitations from engineering requirements to SaaS data constraints to decide if we're the right fit.
5 Silent Killers of AI Agents: Challenge G & Circuit Breakers
Discover the 5 probabilistic failure modes of AI agents (Challenge G) that break DevOps. Learn how Agentic Circuit Breakers stop infinite loops and cost spirals.
Field notes, in your inbox
One email per week. No content calendar — just what we’re building, what broke, and what we changed our minds about.
Operator Stack
The free community where we continue these conversations between posts.
Join free