BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.devopsdays.org//devopsdays-portland-2026//speaker
 //SGN3FM
BEGIN:VTIMEZONE
TZID:US/Pacific
BEGIN:DAYLIGHT
DTSTART:20250908T000000
TZNAME:PDT
TZOFFSETFROM:-0700
TZOFFSETTO:-0700
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251102T020000
RDATE:20261101T020000
TZNAME:PST
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RDATE:20270314T030000
TZNAME:PDT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Build a Platform\, Unleash an Agent on It… and Watch It Burn - M
 ichael Forrester\, Whitney Lee
DTSTART;TZID=US/Pacific:20260908T130000
DTEND;TZID=US/Pacific:20260908T150000
DTSTAMP:20260824T030428Z
UID:pretalx-devopsdays-portland-2026-BYT7TC@talks.devopsdays.org
DESCRIPTION:We built a burrito-ordering AI agent\, put it on Kubernetes wi
 th scoped cluster access and a live terminal\, and let a room of engineers
  try to break it.  It is tons of fun.  \nOver three rounds they push the a
 gent to steal customer data\, deploy malicious workloads\, and take the cl
 uster down. Forkbombs anyone!   \n\nThe first round has nothing protecting
  the agent\, and it usually falls apart in about one prompt. \n\nThe secon
 d runs the same attacks against an ordinary CNCF platform floor\, which bl
 ocks most of them\, though a few run up a bill before they get stopped. \n
 \nIn the third round attendees switch on the remaining guardrails themselv
 es and watch how the agent's behavior and its token cost change.\n\nRunnin
 g it taught us that for most shops\, AI doesn't create just brand-new secu
 rity problems it also makes the ones you already have more expensive. Toke
 n spend becomes its own denial-of-service vector for example. Most of what
  your platform already does holds up against an agent (or for an agent)\, 
 and the gap that's left is smaller and stranger than people assume.  This 
 workshop covers the base guardrails as well as that enhanced gap that you 
 must have for AI workloads.   \n\nWe ran this at AI Engineer World's Fair 
 on June 29th 2026. This version covers what held up\, what broke\, and wha
 t we've changed since.  We have made vast improvements including model tri
 age\, provisioning and sandboxing (as an option).   This will be a very ev
 olved\, much more polished\, and much more expanded presentation than the 
 one we ran at AI Engineer World's Fair. This will be much more hands-on in
 stead of just a web interface. They will actually get access to their own 
 cluster.
LOCATION:Room 327
URL:https://talks.devopsdays.org/devopsdays-portland-2026/talk/BYT7TC/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ozzie and Nova: A Play About Kubernetes Security - Whitney Lee
DTSTART;TZID=US/Pacific:20260909T162000
DTEND;TZID=US/Pacific:20260909T165000
DTSTAMP:20260824T030428Z
UID:pretalx-devopsdays-portland-2026-DT9NDB@talks.devopsdays.org
DESCRIPTION:Ozzie the Overconfident Operator has secured his cluster! He h
 as added Role-Based Access Control (RBAC) and carefully given cluster admi
 n privileges to himself alone. As he self-congratulates\, Nova the Nefario
 us Ne'er-do-well watches from around the corner\, drooling with anticipati
 on. Spoiler alert⎯Ozzie is about to get HACKED.\n\nIn this theatrical pl
 ay\, the speakers perform as Ozzie and Nova\, playfully demonstrating Kube
 rnetes security as Nefarious Nova exploits each of Ozzie's oversights. Wha
 t can Ozzie do when Nova gets the upper hand? How can Ozzie proactively ke
 ep Nova's threats at bay?\n\nYou’ll laugh out loud while discovering clo
 ud native security concepts including Kubernetes network policies\, mutual
  TLS\, admission controller policy enforcement\, runtime security\, git co
 mmit signing\, and multi-factor authentication.\n\nLearn security basics a
 longside Overconfident Ozzie\, who is sure the cluster is COMPLETELY secur
 e this time. There is nothing Nova can do to break... uh-oh. Not again!
LOCATION:Ballroom
URL:https://talks.devopsdays.org/devopsdays-portland-2026/talk/DT9NDB/
END:VEVENT
END:VCALENDAR