Alt, aber immer wieder beeindruckend zu sehen: Arithmetik mit Wörtern. Hier mit unserem GPT-2-Class-Modell und Tokenizer (Credits gehen an @surbaniak, der hat den trainiert) Nanoschnack https://github.com/nanoschnack/nanoschnack/ trainiert nur mit deutschen Texten.

Technisches Detail: dachte es funktioniert nicht. Aber das Leerzeichen vor den Wörter wird mitgelernt beim Tokenizer. Er kennt kein 'Vater', nur ' Vater'.

Mehr am Donnerstag in Heidelberg: https://community.cncf.io/events/details/cncf-cloud-native-heidelberg-presents-building-a-gpt-2-model-from-scratch-nanoschnack/cohost-cloud-native-heidelberg/

Blind automation becomes a threat when enough is at stake.

Kausality adds a causal dimension to Kubernetes controller hierarchies — the who and the why behind every mutation.

#Kubernetes #Crossplane #Terraform #ACK

🎥 youtube.com/watch?v=kfqKqXnRKms

One day you'll have written your last line of code yourself. Write every line like it matters.

Maybe for some of us, that last line already happened.

Challenge accepted 💪

Will talk about my Christmas holiday project NanoSchnack @ Cloud Native Heidelberg meetup, Feb 26.

GPT-2 the Hard Way. Building a LLM from scratch.

https://community.cncf.io/events/details/cncf-cloud-native-heidelberg-presents-building-a-gpt-2-model-from-scratch-nanoschnack/

#transformers #attention #tokens #embeddings #training #pytorch #h100 #inference

It’s enterprise when you port the Claude https://github.com/steveyegge/beads workflow to native Jira. And it works using https://github.com/ankitpokhrel/jira-cli 🎉

I have never commented so much in Jira issues about progress. Well I still don’t but now Claude does for me. Try it.

Newest Claude Code assumes "task" is an internal task, basically shadowing every other system like beads.

This gives same vibes as when Microsoft suddenly shipped Internet Explorer and every other browser had a hard time to get noticed.

Re drift correction with https://github.com/kausality-io/kausality. Here is a mock I want to implement. Directly in @slackhq.

Of course with some "expected drift policy" thing in parallel. This mock is the escalation for those cases slipping through.

Did interview and the candidate directly went to Excalidraw copying my task description from gdoc over. Loved it #visualthinking

Weekend experiment: Kausality

Drift detection for Kubernetes — know when your infrastructure changes unexpectedly.

⚠️ experimental, not ready for production ⚠️

https://github.com/kausality-io/kausality

Multicluster Cert Manager 😎 – Zach contributed the controller sharding implementation to https://github.com/kubernetes-sigs/multicluster-runtime and now uses that to demo how to build a central cert-manager. Awesome use-case.

https://medium.com/@zach.dsmith/a-hub-style-multi-cluster-cert-manager-control-plane-d568ea334998

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.

Hire people who removed code (not tests).
Don't look for people who added the most.