I can proudly say that I’ve been braking production ages before ai was a thing 😎

no, it’s not :) @aarondfrancis explains it well here youtu.be/ZRV3pBuPxEQ?is=e9uh…

reminded me of this meme lol

If you can write the query that proves your data is wrong, run it on a schedule. Then you never need the forensics.

A debugging tool that returns a confident wrong answer is worse than one that returns nothing. Most of the work in writing one is the cases where it must refuse.

very similar to how it was in 2020 with todo list apps

A cloud invoice is a very expensive architecture review that arrives monthly and that nobody reads.

Describe some common Git workflows used in team environments. What are the pros and cons of each?

Nobody notices an 8% drop in average order value. The revenue line still goes up. That is why it survives to month end.

that's how i feel as well

Number of times a client's "unprecedented scale" turned out to exceed one Postgres box: zero. Sample size: not small.

we all have been there maintaining legacy systems. at lease now ai adds comments lol

time to try a better model then buddy

2025 called, they want their meme back

Every data quality incident has the same first hour: someone scrolls through deploys looking for one that sounds plausible.

Yep it needs to be plugged into power though

lol this is too funny 🤣

Explain the different phases of a Kubernetes Pod lifecycle and what happens during each phase.