Greetings from Seoul 🌸🌸🌸🇰🇷

Now back to vacation mode in 🇰🇷

Demo video and code

github.com/carlossg/arc...
github.com/carlossg/arco

For this everything was AI generated, and iterated on. Including the slides or demo video.

For the Google booth I showed a generative websites demo. It showcases how it is already possible to make LLMs build sites live, hyper customized to the user. In this case I showed how it works in Google Cloud, using Gemini, but also Gemma, ollama, Mistral,... in VertexAI

Presentations - YouTube Presentations by Carlos Sanchez

Video is up
youtube.com/playlist?lis...

CNCF-hosted Co-located Events Europe 2026: Beyond Argo Rollouts: Boosting Developer... View more about this event at CNCF-hosted Co-located Events Europe 2026

Beyond Argo Rollouts: Boosting Developer Experience With Intelligent AIOps colocatedeventseu2026.sched.com/event/2DY3U/...

KubeCon was a blast, 12k+ people, and the opportunity to catch up with so many colleagues. One talk at ArgoCon with @kevindubois.com and one demo at the Google booth #kubecon

If you thought the speed of writing code was your problem - you have bigger problems | Debugging Leadership AI coding tools are optimising the wrong thing and nobody wants to hear it. Writing code was already fast. The bottleneck is everything else: unclear requirements, review queues, terrified deploy cult...

This post paints a good view of where we are with AI code generation. The speed of writing code is no longer the problem, and making it faster will not produce results faster
andrewmurphy.io/blog/if-you-...

Playing with Agent Sandbox, interesting k8s project for running agents but also with primitives like warmpool useful for other workloads

Agentic sites: building hyper personalized websites

Demoing Agentic sites on Google Cloud + Adobe AEM at #kubecon, where pages are built by an #LLM just for you

Watching as @kevindubois.com and @csanchez.org getting mobbed after their #ArgoCon demo on how AI can detect and remediate deployment issues with Argo Rollouts! @argoproj.bsky.social

My mental picture how the Agentic Loop should work fully automated for production products. Built on top of existing practices like DevOps and Progressive Delivery. The agents have autonomy but also guardrails to go all the way. Otherwise we just generate code that gets stuck

Goals for #KubeCon next week: learn how other people are most effective with AI, running Agents at scale, how coding fast can become shipping products fast. I have my opinions, but looking forward to yours !

Surely 😃

CNCF-hosted Co-located Events Europe 2026: Beyond Argo Rollouts: Boosting Developer... View more about this event at CNCF-hosted Co-located Events Europe 2026

Next week I'll be at #KubeCon presenting "Beyond Argo Rollouts: Boosting Developer Experience With Intelligent AIOps" with @kevindubois.com at #ArgoCon and demoing some work on generative websites at the Google Cloud booth, come say hi! colocatedeventseu2026.sched.com/event/2DY3U/... #GDE

Ah the joys of traveling. Due to strong winds, my flight MAD-LCG does not even attempt landing in LCG, tries SCQ and finally back to MAD

Playing with 🧲 @CERN 100m underground #VoxxedDaysCERN

Things that happen at #FOSDEM. Met Daniel Stenberg @bagder.mastodon.social.ap.brid.gy the cURL creator

GitHub - zampierilucas/scx_horoscope: Astrological CPU Scheduler Astrological CPU Scheduler. Contribute to zampierilucas/scx_horoscope development by creating an account on GitHub.

Astrological CPU Scheduler. Makes real CPU scheduling decisions based on real-time planetary positions, zodiac signs, and astrological principles. "Because if the universe can influence our lives, why not our CPU scheduling too?" 😆
github.com/zampieriluca...

I'll be speaking at @FOSDEM.org Testing and Continuous Delivery room again next Sunday, showing how to use AI Agents with Kubernetes and Argo Rollouts for Self-Healing Rollouts

GitHub - carlossg/tripit-view: Viewer for tripit data from their GDPR export Viewer for tripit data from their GDPR export. Contribute to carlossg/tripit-view development by creating an account on GitHub.

All of this just took less than an afternoon in total, and with Antigravity browser I could ask Gemini for design changes, fixes, etc and it would go and grab the screenshot to analyze.
Sources: github.com/carlossg/tri...

And then I went down the rabbit hole and added better statistics (flights by year/month/airline), flight visualization (like Jetlovers) and countries visited (like Been), but all automated from the Tripit data.

tripit

You can try it with sample data at tripit.csanchez.org/For your own data you need to ask Tripit for an export of your data (thanks GDPR). Your data never leaves your browser.

Spend some time this holiday building a Tripit data visualization site to play with Google Antigravity. As they killed the API I wanted to ensure I could keep my trip history

I’ve been testing Google Antigravity, and the shift from "Co-pilot" to "Manager" is here. 🚀
Unlike traditional AI editors, Antigravity focuses on autonomous missions. My favorite part? The Browser Integration. It’s a game-changer for visual debugging. blog.csanchez.org/2026/01/15/m...

Another year, too many talks 🙈 @sessionize.com

This was already a problem with low latency dynamic changes (ie. infra updates due to user actions). We solved it by dispatching the changes to the clusters and git at the same time. This could be done with agent instructions but it would be brittle

I was listening to @darinpope.bsky.social @vfarcic.bsky.social about the relevance of GitOps in the age of AI and I still see value on git tracking, but we need a mechanism to merge changes in the cluster made by Agents back to git (or any other backend) and drift detection

Ahu tongariki

This place is surreal
#rapanui #easterisland