Whitney Lee

Software Defined Interviews: Animated Bedtime Stories, Telling Dummies What to Do, and Why Docker Lost, with Viktor Farcic

Kubernetes clusters can carry hidden security issues in infrastructure config, workload settings, RBAC permissions, and containers. Kubescape maintainer Ben Hirschberg joined me on 🌩️Thunder to explain how Kubescape scan... https://whitneylee.com/2026/10/25/kubernetes-clusters-can-carry-hidden.html

[Keynote] DevOpsDays Portland: Ozzie and Nova: A Play About Kubernetes Security

malicious code ≠ vulnerabilities Read the full board notes here: https://gist.github.com/wiggitywhitney/9faced84f4efb2d81cd2538ad9f35d00

AI Engineer World’s Fair: Build a Platform, Unleash an Agent on it…. and Watch it Burn!

Malicious code looks like a normal pull request. And if you merge it, I have bad news for you, friend - you just introduced an active attack. Christoph Hamsen, Senior Security Engineer at Datadog, explains how BewAIre catches attackers' attempts to inject malicious code https://youtu.be/LRW1hI5eduQ

Datadog Illuminated: Rebuilding CI’s Git Serving for the AI Coding Era, with Mike Thompson

When bad data enters a data pipeline, it doesn't always crash something. Sometimes it silently produces wrong numbers, like billing, for months without anyone catching it. Watch the full Datadog Illuminated episode: https://youtu.be/uIuBBp3Ib64

Kagent turns your AI agent into a Kubernetes custom resource, defined declaratively like anything else in your cluster. Watch the full 🌩️Thunder episode: https://youtu.be/z2Xpd-WwmLE #Kagent #Kubernetes #AIAgents #CNCF #CloudNative

Serverless can save you money, but the tradeoff is that it is weird. Your CPU gets frozen between requests. Your code runs in a tiny, isolated sandbox with extremely limited resources. And every millisecond of initialization tim... https://whitneylee.com/2026/10/18/serverless-can-save-you-money.html

OpenLineage's core is intentionally simple. It knows which step in your pipeline ran, and what data it read or wrote. How does it also capture tool-specific details, like a query's runtime or which code version ran? Watch the full video: youtu.be/uIuBBp3Ib64

Every tool in a data pipeline that needs to track where data came from and where it's going has to build a custom integration with every other tool it talks to, so the number of integrations explodes as tools multiply. How does OpenLineage help? Watch the full video: youtu.be/uIuBBp3Ib64

Lin Sun says AI agent security needs a policy layer that covers which tools an agent can access, and whether a human must approve before it acts. Watch the full 🌩️Thunder episode: https://youtu.be/z2Xpd-WwmLE #Kagent #Kubernetes #AIAgents #CNCF #CloudNative

At a high level, OpenLineage is a common observability framework for data pipelines. Watch the full Datadog Illuminated episode: https://youtu.be/uIuBBp3Ib64 #OpenLineage #DataObservability #DataPipelines #DatadogIlluminated

As data moves across teams and orgs, proving it's still correct at every step is what makes it trustworthy. How does OpenLineage capture that proof, every time a pipeline runs? Watch the full Datadog Illuminated episode: https://youtu.be/uIuBBp3Ib64

kagent enables teams to build AI agents that share Kubernetes and platform knowledge across the org. What are the benefits of this? Watch the full 🌩️Thunder episode: https://youtu.be/z2Xpd-WwmLE #Kagent #Kubernetes #AIAgents #CNCF #CloudNative

“The entire world is not HTTP.” In this episode of 🌩️Thunder, Ed Warnicke, Distinguished Engineer at Cisco and Network Service Mesh co-founder, explains how NSM connects individual workloads at Layer 3, regardless of whether they’re ... https://whitneylee.com/2026/10/11/the-entire-world-is-not.html

DevOpsDays Portland: Ozzie and Nova: A Play About Kubernetes Security

Scott Yak's team runs 200 evaluation scenarios against the Datadog MCP server in about 2 minutes, on a local computer. Speedy! How do they do it? Watch the full Datadog Illuminated episode: https://youtu.be/HZ4n1h8j8MY #Datadog #MCP #AIAgents #Evals #DatadogIlluminated

The standard way to evaluate an AI tool: a human writes a question, works out the correct answer by hand, and checks if the AI matches it. Is there a better way? Watch the full Datadog Illuminated episode: https://youtu.be/HZ4n1h8j8MY #Datadog #MCP #AIAgents #Evals #DatadogIlluminated

Nobody on Lin Sun's team wanted to keep losing weekends to networking troubleshooting. What now-popular CNCF technology were they inspired to build? Watch the full 🌩️Thunder episode: https://youtu.be/z2Xpd-WwmLE #Kagent #Kubernetes #AIAgents #CNCF #CloudNative

Datadog's MCP server evals need a constant supply of new questions with correct answers, but those answers go stale fast. How did Scott Yak's team keep generating new evals without writing each one by hand? Watch the full Datadog Illuminated episode: https://youtu.be/HZ4n1h8j8MY

When you build an MCP server, you don’t control the agent calling it or the model powering it. The same server might be called by Claude Code, Cursor, or something completely custom, and it could be powered by anything from a small loca... https://whitneylee.com/2026/10/04/when-you-build-an-mcp.html

DevOpsDays Portland: Ozzie and Nova: A Play About Kubernetes Security

Datadog's Applied AI team tracks eval scores to check how well the Datadog MCP server is performing. If that score stays exactly the same from one week to the next, does that mean nothing changed? Watch the full Datadog Illuminated episode: https://youtu.be/HZ4n1h8j8MY

DevOpsDays Portland: Build a Platform, Unleash an Agent on It… and Watch It Burn

This is the lightboard from my Datadog Illuminated episode with Julien Le Dem on OpenLineage. Before OpenLineage, broken data was nearly impossible to trace back to its source. Read the full board notes here: https://gist.github.com/wiggitywhitney/fc5b325e7706ffcc3f80d7ccdfbd520a

Datadog On Air: Datadog On Air Live From San Francisco Summit 2026

“Managing Kubernetes RBAC manually is a path to burnout.” In this episode of 🌩️Thunder, Saim Safdar, CNCF Ambassador and host of the Cloud Native Podcast, explains how Paralus centralizes Kubernetes access management acr... https://whitneylee.com/2026/09/27/managing-kubernetes-rbac-manually-is.html

Software Defined Interviews: The KubeCon Drone Demo, CFP Reviewing at Scale, and the Agentic AI Foundation, with Lin Sun

@julien.ledem.net, Principal Engineer at Datadog and project lead for OpenLineage, explains how one open standard replaces a tangle of one-off integrations for tracking data across every tool in a pipeline. Watch here: https://youtu.be/uIuBBp3Ib64

InfoQ: AI Meets Internal Developer Platform (InfoQ coverage of KubeCon EU talk with Viktor Farcic)

SPIFFE is the standard for verifying workload identity in production, and SPIRE is the reference implementation that puts it to work. Volkan Özçelik on how far it scales: "It can scale up to millions of nodes, and it still works." Watch the full 🌩️Thunder episode: https://youtu.be/cCHXRzoThsM

With AI, how do we match the speed of trust with the speed of creation? I'm hosting a livestream today! We will be interviewing practitioners from companies like Frame.io, JLL, and Cognition to learn how they're trusting AI to do impactful work. Join here: www.youtube.com/live/pLO-YL2...

Datadog Illuminated: Evaluating an AI-Powered Process That Creates and Tunes Monitors, with Eric Karschner

When a new workload spins up in Kubernetes, it has nothing to prove who it is yet, no credential, no secret. But it can get a verified identity in seconds - how? Watch the full 🌩️Thunder episode: https://youtu.be/cCHXRzoThsM #SPIFFE #SPIRE #WorkloadIdentity #CloudNative

LLMday Austin: The Best Laid Spans: An AI Agent That Instruments Your Code with OpenTelemetry

We secure data at rest. We secure data in transit. But what about data in use? In this episode of 🌩️Thunder, Tobin Feldman-Fitzthum explains how Confidential Containers uses hardware-level trusted execution environments to isolate you... https://whitneylee.com/2026/09/20/we-secure-data-at-rest.html

Datadog Illuminated: Evaluating an Agent That Creates and Tunes Monitors, with Eric Karschner

kagent is a framework to help you make AI agents and run them in Kubernetes declaratively, with YAML. Lin Sun, co-creator of kagent, joined to walk through how it works. Here's the lightboard we made together. Read the notes: https://gist.github.com/wiggitywhitney/97e3347ddc5c00da2b3c07471a2ac277

Lin Sun, Head of Open Source at Solo.io and kagent co-creator, explains how kagent turns AI agent workloads into Kubernetes resources you deploy declaratively, with kubectl apply, alongside everything else in your cluster. Watch the full 🌩️Thunder episode: https://youtu.be/z2Xpd-WwmLE

Volkan Özçelik on workload identity: "I am not identifying hosts because I don't want to say this host can talk to this host. I want to say this application in this host can talk to or cannot talk to this application on the other host." Watch the full 🌩️Thunder episode: https://youtu.be/cCHXRzoThsM

Keycloak's login flow can be enhanced using configurable steps beyond the basic username/password prompt. What are some examples of customized login steps? Watch the full 🌩️Thunder episode: https://youtu.be/Kvb4PMIgR_g #Keycloak #IAM #CloudNative

Volkan Özçelik explains what workload authentication looked like before SPIFFE: "We were baking service tokens to the images, which is not the most secure way." Watch the full 🌩️Thunder episode: https://youtu.be/cCHXRzoThsM #SPIFFE #SPIRE #WorkloadIdentity #CloudNative

Will AI help or hinder a developer’s experience with an internal platform? At KubeCon EU, the audience found out. My bestie Viktor Farcic (Upbound) and I ran an interactive session where attendees voted at key moments on which tools ou... https://whitneylee.com/2026/09/13/will-ai-help-or-hinder.html