Igor Zibarev

@hypnoglow
10 Followers
50 Following
176 Posts
Tech Lead in 💳 payment processing && Head of Go practice @ https://platacard.mx/en · Building a LatAm neobank · talking #fintech, #golang, #ai
🧪 #Claude mad scientist · turning Go bugs into projects nobody asked for
🏴‍☠️ Side quest: weekend #k8s pirate bootstrapping
GitHubhttps://github.com/hypnoglow
Xhttps://x.com/hypnoglow

RE: https://mastodon.social/@stephenblum/115975513491202645

It's wild that we still don't have anything more portable than #bash scripts. And `curl | sh` being the norm says a lot about legacy in our tooling.

RE: https://mastodon.social/@golangch/115977078295734736

Emojis in README: helpful or cringe?

dotfiles/zed/tasks.json at 09da6a85ceaf2a8873130ba3f5ee81b61dc24387 · hypnoglow/dotfiles

:briefcase: My dotfiles [arch, i3, urxvt, macOS, zsh, git, atom, ...] - hypnoglow/dotfiles

GitHub

pushd overlays/dev
kustomize edit set image ...
popd
kustomize build ...

wait, what? is this the way-to-go with #kustomize?

https://github.com/kubernetes-sigs/kustomize/issues/2803

Is #kustomize alive? What are the reasons I should prefer it over #helm?

🏷️ #k8s #kubernetes

Do people still tweak their zsh configs?

I just deleted hundreds of lines from my .zshrc – mostly outdated junk that piled up over the years.

🏷️ #shell #zsh #bash

Running stateful apps in containers is still harder than just running a binary on the host. Haven't seen much progress in this space - any fresh air lately?

🏷️ #docker #containers

If you want your resume to hit - drop the fake precision.

Say what you actually did.
What changed.
What you owned.

"I replaced manual server setup with Terraform" says way more than "improved infrastructure efficiency by 28%".

6(fin)/🧵

What it actually tells me:
This candidate watched a "how to write a great resume" video and followed the template.

Nothing wrong with trying - but fake metrics don't help you stand out.

5/🧵

Or this classic:

> Optimized SQL queries improving performance by 21%.

21% of what exactly? One query? The median? Peak load? Idle time?

That number means absolutely nothing.

4/🧵