Decision tree

Build or buy an AI gateway.

An AI gateway sits between your applications and LLM providers — handling routing, prompt-versioning, evals, guardrails, observability, audit, cost attribution. Build is rarely the answer; this tree walks the conditions where buy / build / skip-entirely actually wins.

4 questions max6 end-states~3-5 minShareable via URL hash

If JavaScript is disabled — the questions in this tree

  1. How many production GenAI use-cases will you run in the next 12 months?
  2. Are you in a regulated industry (financial services, healthcare, government, critical infrastructure)?
  3. Do you have, or are you funding, a GenAI platform team of 3+ engineers?
  4. Do you have hard data-residency / sovereignty constraints that no commercial gateway satisfies?

Re-enable JavaScript to step through interactively. Or jump straight to the related artefacts: diagnostics · reference architectures · writing.

Also on this site