Insights

Notes on building software that lasts

How we think about shipping AI and software that holds up in production — written for practitioners, not press releases.

Agentic AIEnterprise

The state of agentic AI in the enterprise, mid-2026

Agents crossed from demos to production this year. Here's what separates the deployments that stick from the pilots that stall.

Jun 18, 2026 · Trigger Solutions AI Research

RAGEvaluation

Why your RAG system needs an evaluation harness before it needs a better model

Most RAG quality problems aren't model problems. A rigorous evaluation harness finds the real bottleneck — usually retrieval — in days instead of months.

May 27, 2026 · Trigger Solutions AI Engineering

StrategyEnterprise

Build vs. buy for enterprise AI: a decision framework that actually works

The build-vs-buy question is really three questions in disguise. Answer them in order and the decision usually makes itself.

Apr 30, 2026 · Trigger Solutions AI Research