Jeff Hajewski

Reliable AI systems, evals, context, and performance economics.

Writing

This is where I publish reference material on reliable AI systems: evals, context, runtime architecture, monitoring, and performance economics.

No essays are published yet. The archive will start with the first piece that earns its place.

  • The eval pyramid for production agents
  • Context is a systems problem, not a prompt problem
  • What teams misunderstand about AI cost versus latency
  • Why enterprise AI pilots die after the demo
  • Post-deployment monitoring for agentic systems

Suggest a topic by email.