Systems | Development | Analytics | API | Testing

The healthcare industry is buzzing about FHIR (Fast Healthcare Interoperability Resources). Pronounced “fire,” this widely adopted data standard has been revolutionizing how healthcare information is exchanged. But building a truly modern, secure, and scalable digital health platform takes more than just a data standard. Here comes SMART on FHIR, a framework that builds on FHIR to create a universal, secure, and user-friendly way for applications to access electronic health records(EHRs).

Behind every smooth user experience is a maze of APIs quietly handling requests, responses, and data flows. This makes APIs critical connectors that enable applications to communicate and share data seamlessly. When these vital conduits fail, the consequences can be severe—system outages, data breaches, and frustrated users. API testing is the unsung hero ensuring your digital world runs smoothly and securely.

As enterprises increasingly embed AI and Large Language Models (LLMs) into their digital experiences, enforcing robust AI guardrails becomes paramount to safeguard users, protect data, manage operational costs, and comply with regulatory and ethical standards. Think of AI guardrails as essential controls: policy, technical, and operational layers carefully placed around your AI services to detect, prevent, and mitigate any unsafe, abusive, or unintended behaviors.

It’s been almost a year since we released our Konnect Terraform provider. In that time we’ve seen over 300,000 installs, have 1.7 times as many resources available, and have expanded the provider to include data sources to enable federated management of your Konnect organization. There have been many changes in the last year, but there are some changes that we’ve been holding off on as they would break your CI/CD pipelines.

If you build APIs, you’ve probably been in two kinds of conversations. One is the war room, where the team is chasing a latency spike, an outage, or a sudden drop in traffic. The other is the boardroom, where you need to show how the platform is performing, which APIs are growing, and whether your investments are paying off. Dashboards are the tool people turn to in both cases.

Picture this: you’re in the zone — debugging logic, writing endpoints, testing integrations. But suddenly you’re yanked out of flow: copy pasting cURL commands, flipping between environments, chasing down the right API keys, and juggling half a dozen tools just to make a simple change. It’s clunky and error-prone. And honestly, it’s not what you signed up for.

Technical debt is a sneaky hang-up that organizations and teams often ignore until it becomes unmanageable. It creeps in slowly, undermines your team’s productivity, and builds up until your product is hard to maintain and scale. Software systems and development cycles are increasingly being driven and supported by AI, continuous delivery, and cloud-native environments, making it critical to adopt a clear strategy for identifying and reducing technical debt.

Picture this: It's 3 AM, and your phone erupts with alerts. Within minutes, you're drowning in a tsunami of notifications—hundreds of them—while your company's critical services hang by a thread. Your monitoring dashboard looks like a Christmas tree gone wrong, every light blinking red, and you have no idea where to start. Sound familiar?

In the last two parts of this series, we discussed How to Strengthen a ReAct AI Agent with Kong AI Gateway and How to Build a Single-LLM AI Agent with Kong AI Gateway and LangGraph. In this third and final part, we're going to evolve the AI Agent with multiple LLMs and Semantic Routing policies across them. In this blog post, we'll also explore new capabilities introduced in Kong AI Gateway 3.11 that support other GenAI infrastructures.

In my previous post, we discussed how we can implement a basic AI Agent with Kong AI Gateway. In part two of this series, we're going to review LangGraph fundamentals, rewrite the AI Agent and explore how Kong AI Gateway can be used to protect an LLM infrastructure as well as external functions.