“We go through our Kubernetes clusters and we’re killing nodes every 2 minutes in production constantly.”
At Google Cloud Next, I hosted a fireside chat with David Faircloth (VP of Architecture and Engineering at Wendy’s) about how Wendy’s scaled to 99.95% availability in five years.
Wendy’s built their way there by designing systems to fail gracefully. Observability is what made that strategy visible, and chaos engineering is what proved it out in production.
“I just Googled Kubernetes and I know more than you do.”
An investor actually said that to Emily Long’s all-woman founding team during a pitch!
In this episode of Software Defined Interviews, Emily Long, CEO and co-founder of Edera, talks about building a hardened container runtime that secures infrastructure foundations instead of chasing detect-and-respond alerts, and what it’s like raising a deep-tech Series A when investors have opinions like that one.
At Observability Day, KubeCon EU, I presented Spinybacked Orbweaver, an AI agent that automatically instruments your JavaScript code with OpenTelemetry using OTel Weaver as the standards contract.
“I’ve done migrations similar to this before we had these kind of tools and it was a very different world.”
In this episode of Datadog Illuminated, Nick Isaacs, engineer on Datadog’s AI DevX team, walks through the LLM-powered workflow his team built to migrate thousands of HTTP endpoints out of Datadog’s webapp monolith into microservices.
The system pairs Claude Code for transformation with GPT-5 for code review, orchestrated by LangGraph, with a self-correcting verification loop that runs the type checker, tests, and server start and fixes its own errors before a human sees the PR.
“When I talk to people from projects who are really struggling with this, it’s because they never had systems to deal with low-quality contributions to begin with.”
In this episode of Software Defined Interviews, Josh Berkus, Kubernetes community architect at Red Hat, explains why AI-generated contributions haven’t created a crisis for Kubernetes, even though they have for other projects.
Spoiler alert: projects that invested in contributor triage, automated tests, and onboarding years ago are absorbing the AI-contribution onslaught.
Humans have had identity solutions for years. For example, passwords, two-factor auth, and SSO. But what about services? They need identity too, especially now, with AI agents in the mix.
In this 🌩️Thunder episode, Henry Avetisyan, Software Engineer at Yahoo and lead maintainer of Athenz, explains how Athenz issues every workload a short-lived certificate as its identity, valid for mere hours, and how that cert gets refreshed automatically.
When something is compromised, you stop renewing. The certificate expires on its own.
In this episode of Software Defined Interviews, Heidi Waterhouse, co-author of Progressive Delivery, explains why the DevOps loop is missing a critical piece: user feedback.
Software should fit to our hands, not force us to contort ourselves to fit to the software.
“In Kubernetes, ‘pod’ refers to both a configuration stored in etcd AND the actual running container.”
In this episode of 🌩️Thunder, Abby Bangser (Principal Engineer at Syntasso) and I untangle CRDs, custom resources, controllers, and operators. These terms can get used interchangeably, but they mean very different things.
A CRD is a schema. A custom resource is data in etcd. A controller is what does the work.
This is the finished lightboard from the 🌩️Thunder episode on Keycloak. Ever wonder why it’s so easy to plug an app into Keycloak? This board explains the OpenID Connect flow, step by step.
In this episode of 🌩️Thunder, Duffie Cooley, Field CTO at Isovalent and CNCF TOC member, explains why Cilium does far more than a traditional CNI.
Cilium attaches an eBPF program to every pod’s network namespace at creation time. That single hook is how it delivers networking, security, and observability at native kernel speed.
“AI’s role in software development is primarily that of an amplifier.”
In this episode of Software Defined Interviews, Nathen Harvey, who leads the DORA research program at Google Cloud, explains what 15 years of research data reveals about AI’s impact on software delivery.
During my sunrise hike today, I spotted six sneaky Underwing Moths on a single tree.
Can you find them all? ♫
On 🌩️Thunder, Kubescape maintainer Ben Hirschberg shows how the tool catches misconfigurations, then watches runtime behavior with eBPF & turns that into a least-privileged NetworkPolicy proposal.
Your GenAI feature returned a 200. The model did indeed return text to the user. But was the answer any good?
In this 🌩️ Thunder episode, Austin Parker (Director of Open Source at Honeycomb) explains why traditional observability falls short for GenAI. With deterministic software, success is binary: it worked or it didn’t. With GenAI, a successful HTTP response tells you nothing about whether users are actually happy.
Austin covers how OpenTelemetry’s semantic conventions let you combine user feedback (thumbs up, thumbs down) with performance data (token counts, model, cost) in the same telemetry. He also talks about “architectural blindness” in coding agents and how feeding trace data back into your agent helps it understand your system’s execution paths.
