Systems | Development | Analytics | API | Testing

One API request took 302 milliseconds. Nothing failed. CPU was mostly idle. The response was correct. The trace made the problem obvious: eight inventory calls, each waiting for the previous one. But the trace could not tell me why the application made eight calls, or whether changing their execution would preserve the response. It showed the shape of the wait, not the input that created it.

There’s something satisfying about contract testing the contract-testing tool. The SmartBear MCP Server is the integration layer between AI coding assistants and the PactFlow API, so when we decided it needed Pact consumer tests of its own, we were subjecting our own code to the same standards that we recommend.

Every team running production LLMs has had the same idea: not every request needs the frontier model. Intelligent model routing (or LLM routing)— choosing a model per request on criteria such as task complexity, cost, latency, or quality — enables more efficient model usage.

The pull request looked perfect. Two approvals, clean diff, all checks green. The team shipped it Thursday afternoon. By Friday morning, support tickets were coming in. On a popular mid-range Android device, the new checkout button rendered behind a promotional banner: visually present but physically untappable. Every reviewer read that code, yet none of them could have caught it, because the defect was never visible in the diff. It only ever appeared on the device.

Many teams assume that if an API passes functional tests, it’s ready for production. Functional testing, however, only verifies that endpoints return correct results for individual requests. It does not reveal how the API behaves when subjected to heavy traffic or sustained load. Under real-world conditions – such as a surge of users logging in simultaneously – APIs that pass functional checks may still experience latency, errors, or outages.

You've probably seen the demos. An AI model kicks off a task, hands pieces of it to other AI models, and somehow the whole thing gets done. Emails drafted, code reviewed, reports summarized — all without a human in the loop. While a single agent doing one thing is impressive, the true paradigm shift occurs when transitioning from single-agent to multi-agent AI systems. It looks like magic. It isn't.

You Can't Govern What You Can't See A decade ago, cloud and API sprawl got ahead of governance, and enterprises spent years trying to account for costs they'd never tracked. Today, we're seeing the same pattern around AI, with hundreds of customers proxying traffic via Kong AI Gateway, which includes LLM, MCP, and agent connectivity. *AI spending will reach $2.59 trillion in 2026.* I regularly like to share what we're seeing in production at Kong.