Issue #78

The AI tool you tried last quarter waited for a prompt, hallucinated a number, then asked if you'd like a summary.

Viktor opened a PR at 2am, rebased it against main, ran your test suite, and posted a note in #eng: "Two flaky tests in payments service, both pre-existing. Recommended merging after fixing them." Then drafted the customer reply for the support ticket the bug created.

That's 619K autonomous actions per day across 20,000+ teams. Not chat replies. Real work shipped to GitHub, Stripe, Linear, Notion, and 3,000+ other tools, from inside Slack and Microsoft Teams.

You don't supervise him any more than you supervise a senior engineer.

SOC 2 certified. Your data never trains models.

"It's what you probably originally thought AI was going to be when you first heard of it in sci-fi movies." Tyler, CEO.

Get Started for Free

If you're building on MCP, the next SDK release is one you won't want to overlook. Felix Weinberger previews the upcoming beta SDKs, introducing the changes needed to support the biggest evolution of the Model Context Protocol to date. With major protocol updates on the horizon, developers have an opportunity to prepare their applications before the new specification becomes the standard.

Do you really need to rebuild your APIs for AI agents? Christian Posta explores a new approach that transforms existing OpenAPI specifications into MCP-compatible tools, allowing organisations to make their APIs agent-ready without starting from scratch. It's a practical example of how enterprises can bridge today's API investments with tomorrow's AI-driven workflows.

Adding more MCP servers doesn't automatically make AI agents more capable. MCP Pulse explores practical strategies for getting the most out of MCP, from choosing the right tools and structuring workflows to avoiding common pitfalls that reduce performance. As the MCP ecosystem grows, success will depend less on the number of integrations and more on how thoughtfully they're used.

AI can create APIs in seconds but speed without governance can quickly become a liability. Daisy Priya explores why AI coding assistants and API governance platforms play complementary roles, arguing that as AI accelerates API development, organisations need stronger controls to ensure consistency, security, and compliance. The faster AI generates APIs, the more important governance becomes.

What happens when a government, not your vendor decides your AI model can no longer be used? Using Anthropic's recent model shutdown as a case study, this article explores why AI sovereignty is rapidly becoming a boardroom concern. It argues that resilience isn't just about choosing the right model—it's about designing architectures that can withstand regulatory, commercial, and operational disruption.

Great API documentation isn't just for developers anymore it's increasingly being read by AI. The Readme team explores what it takes to make API documentation "LLM-ready", from clearer structure and richer examples to AI-friendly formats that help coding assistants and agents understand your APIs accurately. As AI becomes a primary consumer of technical documentation, documentation quality is taking on a whole new level of importance.

Your API specification may pass every linting check but would it actually work in production? This presentation by Specmatic explores a blind spot in traditional OpenAPI linting, arguing that syntactically valid specifications can still contain semantic errors that lead to runtime failures. It introduces a different approach that focuses on whether an API specification is truly executable, not just technically valid.

The Ikenna Consulting Newsletter curates intelligence on the trends, technologies, and practices shaping API platforms.