Systems | Development | Analytics | API | Testing

The context graph — not the UI layer or system of record — is the true competitive IP of the AI era, and Kong built Context Mesh to help companies govern it. Without the right context layer, AI agents are generic and interchangeable regardless of which LLM is underneath. Companies that own and protect their context graph can differentiate their agentic workflows; those that don't are left with legacy CRUD backends that don't translate to agentic use cases. Context Mesh gives enterprises policy and governance over what agents can consume — the rulebook for all context flowing in and out.#Shorts.

Most AI agents are generalists that struggle with the nuances of streaming data and Kafka infrastructure, often leading you down a rabbit hole with many tokens spent. In this session, live from Current London, we’ll show you how to close the context gap with Skills for Kafka: structured files that are playbooks and level-ups for agents to handle complex multi-step tasks like reviewing schema changes, critiquing DLQ policies and auditing topics for best practice. We will walk through how these skills work, demonstrate the different ways to use them (including with MCP servers) in Cursor and Claude.

Is your organization actually driving value with AI, or just scratching the surface? @SPGlobalMarketIntelligence's Saugata Saha joins to share the 4-part journey to real AI maturity. He warns that without a deliberate action plan to completely transform day-to-day employee workflows, any productivity gains will just be "squishy" metrics that fail to impact the bottom line. Head over to your favorite listening platform to catch the full episode!

Visit: Speedscale.com to learn more.

Stop wasting money on AI API tokens in your local, test, and CI environments. In this demo, I show how to record real LLM responses and replay them for free using proxymock and stop surprise bills.