EngineeringFrameworks
Battle-tested methodologies for building software in the age of AI agents. The artifacts boards ask for and teams actually use.
The Library
Six frameworks, one operating model.
Each framework answers a specific question I hear from engineering leaders. Read them standalone — or use them together as the backbone of an AI transformation.
AI Governance Framework
Policy, risk, and compliance that enables velocity
A board-ready governance model covering usage policy, model approval, data boundaries, vendor risk, and incident response. Pair with the public governance checklist to assess current state.
For: Leaders facing audit, compliance, or board questions
Build vs. Buy Decision Matrix
The #1 question on every CTO's desk
A structured decision framework for evaluating AI tooling, platforms, and foundation models — weighing switching cost, lock-in, data sovereignty, and time-to-value against internal capability.
For: CTOs navigating vendor sprawl and foundation-model choice
AI Architecture Decision Records
Auditable decisions for AI-native systems
An ADR template tuned for AI architecture: model selection, retrieval strategy, evaluation harness, guardrails, and observability. Decisions that survive team turnover and post-incident reviews.
For: Platform teams standardizing AI infrastructure patterns
LLM Cost & Evaluation Framework
Answer "what is this actually costing us?"
A unit-economics and evaluation model for production AI — per-request cost, eval coverage, regression gates, and model-swap decisions. The numbers a board expects you to have.
For: Teams operating LLM features at meaningful scale
Stay Current
New frameworks ship to the newsletter first
Subscribe to get each framework as it's published, plus the field notes behind it — real engagements, real tradeoffs.
Subscribe for UpdatesKeep reading
More on the way in the newsletter.
Short, focused dispatches on shipping AI safely at scale. The thinking before it turns into a playbook chapter.