Andrew Odendaal

This is the index for the four programming puzzle problems on this site.

I’d been running Cluster Autoscaler on our production EKS cluster for years.

The challenge Find the longest substring of a string where characters appear in non-decreasing alphabetical order.

The challenge Find the longest substring of a string where characters are in non-decreasing alphabetical order.

The challenge Given a lowercase string of letters (no spaces, no digits), return the length of the longest substring made only of vowels (aeiou).

The challenge Given a lowercase string of letters (no spaces, no digits), return the length of the longest substring that contains only vowels (aeiou).

The challenge Given a lowercase word, return every permutation of its letters in alphabetical order.

The challenge Given a word (a string of lowercase letters), return all its permutations in alphabetical order.

The challenge Given a slice of positive integers, return the value with the most digits.

The challenge Given an array of positive integers, return the number with the most digits.

The challenge Given an array of positive integers, return the number with the most digits.

eBPF is the biggest shift in Linux observability since strace.

Single-account AWS is a ticking time bomb.

Security as a gate at the end of the pipeline is security theater.

Tokio is Rust’s killer app for network services.

You don’t know your system is resilient until you’ve broken it on purpose.

I’m going to say something that’ll upset a lot of people: pandas had its run.

I’ve spent years writing Python for DevOps tooling and Go for services.

Gateway API is what Ingress should have been from day one.

Bedrock is AWS finally getting AI right.

If you’re not running scheduled terraform plan, you have drift.

Everything I’ve learned building on AWS since 2012, organized by domain.

This is the hub for everything I’ve written about Kubernetes.

I’ve been running Kubernetes in production for years now, and there’s a specific kind of pain that only hits you once you cross the threshold from “a couple of clusters” to “wait, how many do we have again?”

Last year I ported an image processing pipeline from JavaScript to Rust compiled to WebAssembly.

Aurora Serverless v2 is what v1 should have been.

99.99% availability sounds great until you realize that’s 4 minutes and 19 seconds of downtime per month.

I mass-deleted requirements.txt files from a monorepo last month.

NGINX Ingress is the Honda Civic of ingress controllers.

I deleted roughly 2,000 lines of orchestration code from our payment processing service last year.

Most Terraform code has zero tests.

I spent four hours on a Tuesday night debugging a 30-second API call.

If you’re not scanning container images before they hit production, it’s only a matter of time before something ugly shows up in your environment.

EventBridge is the most underused AWS service.

Don’t optimize until you’ve profiled.

Operator SDK vs kubebuilder — I pick kubebuilder every time.

I got paged at 3am on a Tuesday because a Rust service I’d deployed two weeks earlier crashed hard.

I use both.

Platform engineering is DevOps done right.

CPU-based autoscaling is a lie for most web services.

Goroutines are cheap.

VPNs are not zero trust.

If you’re writing Python without type hints in 2026, you’re making life harder for everyone — including future you.

I got a call from a startup founder last year.

I’m going to say something that’ll upset people: if your developers have cluster-admin access in production, you’re running on borrowed time.

I once inherited a project with a single main.tf that was over 3,000 lines long.

ArgoCD won the GitOps war.

I started learning Rust as someone who’d spent years writing Python scripts and Go services for cloud infrastructure.

ECS is underrated.

I’ve shipped Docker images to production for years now, and the single biggest improvement I’ve made wasn’t some fancy orchestration tool or a new CI platform.