In partnership with
Issue #80
Stop making AI decisions in the dark.
Leadership is asking: are we getting value from AI? Which tools are worth the spend? Where are we exposed? Right now, most teams have no idea.
Harmonic Security Usage Explorer changes that.
You get a complete picture of how your organization uses AI, automatically categorized into custom tasks and use cases.
You’ll see the projects being worked on, who’s using what tools, where AI investments are driving value, and where employees are engaging in risky behavior.
CIOs can rationalize spending and cut wasted licenses. CISOs can pinpoint where risk exists and neutralize it. AI committees can show exactly how their efforts are paying off.
Table of Contents
Introduction
In my latest podcast episode, "Who certifies the APIs your agents are calling?” Nicholas from Apyhub argues for the need for dynamic, machine readable API audits. Check out a clip, and the full episode.
Everyone is talking about AI agents. Few are asking whether the digital foundations are ready. ApiShare argues that successful AI adoption starts long before deploying agents. Organisations must first treat APIs and other digital assets as governed digital products—complete with ownership, lifecycle management, discoverability, and policy enforcement. Without that foundation, AI agents simply amplify existing complexity and risk. ApiShare also introduces AI-powered design assistance, integrated governance, and lifecycle automation that shift governance earlier into the API development process.
What Happens When AI Becomes Your Best Developer?
Writing code is no longer the benchmark. Solving the hardest engineering problems is. This behind-the-scenes case study explores how Cursor is using Anthropic's latest models to tackle the most challenging software engineering tasks, revealing what separates frontier AI from everyday coding assistants. It offers an early glimpse into how AI is changing not just developer productivity, but the nature of software engineering itself.
The Best MCP Servers of 2026:Which Ones Are Actually Worth Using?
With thousands of MCP servers now available, choosing the right ones has become harder than building with them. Elena Kovacs cuts through the noise with a curated selection of the most useful MCP servers across development, productivity, search, databases, and automation. Rather than listing every option, it focuses on the tools that deliver real value and explains where each one fits in a modern AI workflow.
Most API Attacks Don't Break In, They Log In
What if your biggest API security threat already has valid credentials? J Simpson reveals why authenticated traffic can be far more dangerous than anonymous attacks, highlighting the behavioural patterns that distinguish legitimate users from malicious actors. As AI agents and automated systems generate increasingly convincing API traffic, organisations need to look beyond authentication and start analysing behaviour.
Will AI Agents Replace Traditional APIs? Not Even Close.
If AI agents speak MCP, do APIs still matter? James Hirst tackles one of the biggest misconceptions in the agentic AI era, arguing that MCP doesn't replace APIs, it builds on them. While AI agents may consume services through MCP, the underlying business logic, data, security, and governance still live within traditional APIs. As organisations adopt AI, the challenge isn't choosing one or the other—it's governing both together.
AI Costs Are Rising, Can You Control Them Before the Bill Arrives?
As AI agents become part of everyday development, keeping track of token usage and costs is becoming a business challenge, not just a technical one.
Nolan Di Mare Sullivan of Speakeasy introduces AI Cost Control, giving organisations visibility into AI spending across models, teams, and agents while helping to identify runaway workloads before they become expensive surprises. It highlights why cost governance is quickly becoming a core capability of enterprise AI platforms.
In the AI Era, Winning the Category Isn't Enough
When developers ask AI for recommendations, they're not searching by category—they're describing a problem. Adam DuVander argues that technical products gain visibility by owning the context in which they're used, rather than competing on broad category labels alone. As AI assistants increasingly shape how developers discover tools, success depends on creating content and documentation that answers real-world use cases, not just product comparisons.
The GraphQL Performance Problem Every Team Eventually Faces
GraphQL makes data fetching easier, but it can quietly introduce one of the biggest performance bottlenecks in modern APIs. Brendan Bondurant explores the N+1 query problem, explaining why seemingly simple GraphQL queries can trigger an explosion of backend requests as applications scale. More importantly, it looks at practical strategies for eliminating the issue before it impacts performance, reliability, and infrastructure costs.
What do you think of this newsletter issue?
About this newsletter
The Ikenna Consulting Newsletter curates intelligence on the trends, technologies, and practices shaping API platforms.
