AI SDLC · 2 min read
What an AI SDLC actually looks like
Adding a coding assistant to an old process is not an AI SDLC. Here is the lifecycle we run instead, stage by stage.
Insights
From the build log.
What we learn building enterprise platforms with AI engineers — the lifecycle, the harness and the products.
8 articles shown
Harness engineering · 2 min read
Harness engineering: the discipline that makes AI code shippable
The model is the engine. The harness — context, tools, guardrails and checks — decides whether it gets you anywhere.
Harness engineering · 2 min read
Why every AI-written commit needs a regression harness
AI makes code cheap. It does not make correctness cheap. A regression harness is how you keep the second from slipping while you enjoy the first.
AI SDLC · 2 min read
Spec-first: writing requirements an agent can build from
The quality of AI-built software is set before a line of code is written. A practical template for specs that agents follow and harnesses can check.
Commerce · 2 min read
Agent-ready commerce: letting buyers’ AI shop through MCP
Your next customer may be an AI assistant acting for a person or a procurement team. Here is what it takes to sell to one.
SaaS infrastructure · 2 min read
Multi-tenant SaaS on Google Cloud: lessons from building Shipyard
Tenancy, billing and provisioning are the parts of SaaS everyone rebuilds and few enjoy. What we learned turning them into a platform.
Build log · 1 min read
The one-founder engineering firm
How a single senior engineer, directing AI virtual engineers with real discipline, builds and runs a portfolio of enterprise platforms.
Security · 1 min read
Security in the AI SDLC: the checks we never skip
AI agents write code faster than anyone can eyeball it. These are the controls that keep speed from becoming exposure.