Blog
The AI Engineering Playbook
Practical thinking about AI engineering, building production systems, and making dev teams better with AI. No hype. No jargon walls. Just useful ideas from the work we do every day.
You Gave Your AI Agent Keys You'd Never Give an Intern
Stop trying to win the prompt-injection war. Your real agent-security problem is a non-human identity holding a master key with the judgment of a new hire. Here's how least agency fixes it.
10 min read
Build, Buy, or Partner: The AI Agent Decision Just Changed
Anthropic, Google, and OpenAI just turned the agent harness into a commodity. Here's what to build, what to rent, and where a partner actually fits.
10 min read
Why Most Multi-Agent Systems Won't Make It to 2027
96% of organizations are using AI agents. Gartner predicts 40%+ will be canceled by 2027. Here's what's killing them in production, and the patterns that ship.
11 min read
Your AI Is Live. Now What? A Field Guide to LLM Observability
89% of teams with agents in production have observability set up. Only 52% run evals. Here's why that gap is a problem, and what the full stack actually looks like.
12 min read
How to Measure AI ROI Without Lying to Yourself
49% of organizations can't demonstrate the value of their AI investments. The problem isn't the AI. It's how they're measuring it.
10 min read
The Model Price Collapse: What to Build Now That You Couldn't Six Months Ago
Frontier models just got 67% cheaper. Here's what that actually means for the economics of building with AI.
8 min readHave a question about AI?
Whether something here sparked an idea or you're working through an AI challenge, we're always happy to talk it through.