Sharing AI practices across a team
Build a team CLI for shared AI skills and tools
A hands-on Go example that installs team-owned skills and rules, configures coding agents, and runs a useful local MCP server.
Field notes by David Hayes
What I’ve learned building systems people depend on.
I write about architecture, distributed systems, modernization, and technical leadership. The examples come from systems I have built, teams I have led, and lessons production makes impossible to ignore.
AI-native engineering
AI is changing how I build and raising the bar for engineering.
I use coding agents to explore faster, automate mechanical work, and expand what teams can deliver. My writing focuses on turning that leverage into reliable production software through better architecture, evidence, and human judgment.
Explore my writing on AIAI Works Best When Engineers Lead
AI gives engineers more reach when people set purpose, direct the work, exercise judgment, and own what happens next.
Featured writing
Architecture that survives contact with production
Working implementations and field notes on modeling business rules, testing failure modes, and preparing teams to operate what they ship.
The work is people, goals, and decisions
Architecture is technical system design — boundaries, data, contracts, failure, rollout — plus the leadership to align people on those choices against a measurable goal.
Build your own webhook transport
How to build a webhook registry and delivery system on Postgres using outbox rows, delivery attempts, signatures, retries, and inbound inbox rows.
Domain-Driven Design in TypeScript
Protect business rules with aggregates and value objects
Learn how tactical DDD uses an aggregate root, entities, and immutable value objects to keep an enterprise seat-allocation rule true.
Production controls
Lease renewal, fencing, graceful drain, and housekeeping for Postgres-backed durable work after order:9182 ships.
Designing an asynchronous boundary that survives outages
A team case study in replacing a brittle synchronous integration with durable intent, idempotent delivery, explicit recovery, and operational evidence.
Production readiness is a design discipline
A practical team standard that connects architecture, automated evidence, observability, rollout, and incident recovery.
Deep technical series
Follow the decisions, not just the result
Long-form, code-backed guides that trace architectural decisions from constraints and tradeoffs through implementation and operations.
Durable Work in Postgres
A practical path from one Postgres claim loop to recovery, idempotency, competing workers, stable ownership, and cross-service handoffs.
Domain-Driven Design in TypeScript
Learn strategic and tactical Domain-Driven Design by working through one non-trivial enterprise SaaS onboarding domain in TypeScript. The case study is the teaching vehicle; DDD is the subject.
Sharing AI practices across a team
A hands-on example of how one team might collect, review, share, and improve skills, rules, MCP configuration, and other useful AI practices together.
Architecture Is a Team Sport
Software architecture as technical system design — boundaries, data, contracts, failure, rollout — practiced through leadership, communication, and collaboration aimed at measurable business goals.
Topics
Latest
Recent posts
- Architecture Is a Team SportGet knowledge in before the plan hardensPart 2
- Architecture Is a Team SportThe work is people, goals, and decisionsPart 1
- CouchDB data modeling: what belongs in a document
- Sharing AI practices across a teamBuild a team CLI for shared AI skills and toolsPart 1
- AI Works Best When Engineers Lead
- How to review AI-assisted code before it ships
- Domain-Driven Design in TypeScriptPut strategic and tactical DDD together in TypeScriptPart 6
- Build your own webhook transport