This Tuesday started like a Monday.

Or, as Peter Steele would sing:

“I don’t wanna be me.”

Gym pass didn’t work. Leasing guy gaslit me with contract dates.
QA?
QA gets a last-minute handoff at the end of the sprint with a cheerful “will you manage?”
And that’s not a question – that’s a stress test. 🔥

And then, the ticket shows up.

“We just need a thousands separator.”
Said someone, not realizing how not just that was.

Turns out – the separator shows up only on the banner. Everything else?

“Not in scope.”
“We just decided it’ll be banner-only.”
“We’ll add the rest later.”
“That’ll be in a different ticket.” 🔥

What does the ticket even look like?

Two blurry screenshots and poetic license.
It reads like a haiku written at 1am — drunk and half-blind. And any actual requirements – if they exist – are buried somewhere in a 128-message-long dev chat thread.
If you weren’t online for that thread live, good luck guessing what was really agreed on. 🔥

Meanwhile, there’s a Big QA Milestone happening.

Not a release milestone – a “Let’s Get Serious About Quality™” kind of milestone:

90% E2E test coverage,
test docs written before merging,
QA linters,
and CI pipelines triggering our tests in every dev repo like a holy sourdough starter. 🔥

It was supposed to be a breakthrough moment.

Testing, not as an afterthought, but as a condition for existence.

And I – the last QA heretic – quietly suggest:

“Smoke, sanity, ROI. Automate what breaks the most.
E2E only where it’s critical.” 🔥

And then it hits me.

  • We have a backend – that changes every sprint.
  • A CMS – maintained by a different team.
  • A user-service – separate backend, separate team.
  • And it all ties together… on the frontend.

The system is distributed, mutating, and alive.
And we’re aiming for 90% E2E coverage on a thing that’s half unknown and half unreachable.

QA in 2025: trying to ensure quality in a system that speaks through headsets, but no one’s listening. 🔥

And when QA dares mention sanity tests, smoke, or ROI?

“Let’s just focus on coverage and regression trends…”

The QA team ramps up – because business wants metrics.
So there will be metrics. There will be dashboards. There will be charts.
A bit forced, if you ask me.
Kind of just to keep business happy. 🔥

And me?
I just want to do it right.
Not spend months fixing flaky E2E.
Not write a test that dies before the ticket does.
Just build tests that make sense.

But the only response I get is a few sighs and eventually the classic:

“We’ll just do it our way.” 🔥

All of this wrapped in a beautifully empty buzzword:

“Interdisciplinary team.”

In reality?
Four teams. Four priorities. Four sprints. Zero shared direction.
And QA is left to stitch it all together like Frankenstein – with a deadline. 🔥

And honestly…

Are we talking about application sanity – or my own?

After a few sprints full of ticket mutations, 90% E2E targets, and reminders that smoke is “not a cigarette,” you start wondering who really needs a sanity check.

Spoiler: QA. Always QA.

PS. QA Therapy – Tuesday Edition

You know what else?

I spent over an hour today comparing fonts, colors, spacing – element by element – to the Figma design.
Because someone said, “but it works.”
And I was the idiot who had to prove it doesn’t look like it’s supposed to.
That’s not testing.
That’s UI babysitting. 🔥

And to top it off – today I got bit in the ass by the “interdisciplinary” dream.

They got the info on Thursday.
On Friday they replied:

“We’ll deliver it Monday.”
Monday… nothing.
Tuesday?
A whole lotta nothing. Just a bannerless void. 🔥

No data. No way to test.
And there I was, test in hand, waiting for something that should have been ready days ago.

“Interdisciplinary”?
More like intervention-needed QA.
Because once again, I’m the one patching what no one delivered. 🔥

And that’s why this post had to be written.
Not as a manifesto. Not as a complaint.
But as a QA battle journal – for everyone who says “sanity” and hears only silence in return. 🔥