API failures often come from missing lifecycle controls rather than difficult technology. These generalised stories from my experience show how unchecked assumptions around versioning, testing, logging, and access control can create avoidable problems.
Contact
Let’s talk about APIs
I’m an Architect for IBM API Connect SaaS, with nearly twenty years in integration. I started in operations and SRE before moving into architecture, and I’m interested in how APIs are designed, how systems behave under pressure, and how to make the reliable path the easy one.
Get in touch about API architecture, integration, SRE, speaking, a project, or simply to compare notes.
Send me a message
I read everything that comes through, and I’m happy to talk about APIs, API architecture, integration, SRE work or the tools I’ve built around them.
Hosted on Letterbird
Recent writing on APIs
A few recent posts tagged api, covering design, observability and reliability.
- 6 Oct 2026
technicalAPI failure stories and what they taught me - 13 Apr 2026
technical - 28 Mar 2026
technicalObservability as a design requirementObservability is a design decision, not an afterthought. From the platform team to API consumers to AI agents, knowing what is happening across your API — and building that in from the start — is what gives you confidence when things go wrong in production.
- 14 Mar 2026
technicalDesigning for Multi-Region API DeploymentsA practical guide to designing multi-region API deployments - covering data consistency, auth dependencies, traffic routing, and failure modes, with real-world lessons from IBM API Connect SaaS.
