QA Wolf (@QAWolfHQ) / X

QA Wolf

237 posts

QA Wolf

@QAWolfHQ

QA Wolf is zero-effort automated QA. We get you to 80% coverage in four months — and keep you there — so your team can stay focused on shipping.

Seattle, WA

Joined July 2019

QA Wolf
@QAWolfHQ
Jun 5
Your QA lead should be feral, ungovernable, haunted by edge cases. Known to file 47 bugs on a login button and call it a light day.
Teddy Blank
@teddypowday
Jun 3
Your CEO should be strong. Your CTO should be wise. Your COO should be wicked, cunning, of mysterious origins, fluent in the dark arts, blurry in pictures,
148
QA Wolf reposted
QA Wolf
@QAWolfHQ
Jun 4
AI made building internal tools feel like the obvious call. A Claude subscription and a few afternoons — why pay for a tool when you can have something running by Thursday? The problem: the build side of the ledger got cheaper. The own side didn't.
126
QA Wolf
@QAWolfHQ
Jun 4
Replying to @QAWolfHQ
The engineers were working with incomplete information and tools that feel like magic sometimes. But buying the right tool and pointing builders at problems only your team can solve is often the smarter play.
The Graveyard of AI-Built Internal Tools Keeps Growing: So, Why Are Teams Still Deciding to Build?...
From qawolf.com
14
QA Wolf
@QAWolfHQ
Jun 4
Replying to @QAWolfHQ
Before you build, answer these honestly: → Who owns this in two years? → What does the team give up to build this? → What's the exit strategy if it fails? If any of those don't have a clear answer, you have your answer.
18
QA Wolf
@QAWolfHQ
Jun 4
Replying to @QAWolfHQ
Three things driving bad build decisions: - The AI confidence effect ("the hard part is done" after the demo) - The customization illusion (your requirements aren't as unique as you think) - The cost illusion (you calculated the sprint cost, not the ownership cost)
31
QA Wolf
@QAWolfHQ
Jun 4
Replying to @QAWolfHQ
Here's the pattern: Team decides to build. Prototype looks great. Someone writes an engineering blog post. People are inspired. Six months later, they're calling the vendor back. The post stays up. The lesson doesn't spread. Read the post:
The Graveyard of AI-Built Internal Tools Keeps Growing: So, Why Are Teams Still Deciding to Build?...
From qawolf.com
43
QA Wolf
@QAWolfHQ
Jun 4
AI made building internal tools feel like the obvious call. A Claude subscription and a few afternoons — why pay for a tool when you can have something running by Thursday? The problem: the build side of the ledger got cheaper. The own side didn't.
126
QA Wolf
@QAWolfHQ
May 27
Customer: Your app is broken. How the customer was using your app:
Just Dave now
@justdavenow89
May 27
This dog is like fuck you I wont play by your rules
00:00
153
QA Wolf reposted
QA Wolf
@QAWolfHQ
May 26
Everyone's talking about how AI is creating 10x engineers (or 100x engineers). Nobody's talking about what it's creating on the other end of the spectrum. We wrote about the -10x engineer. A thread 🧵
182
QA Wolf
@QAWolfHQ
May 26
Replying to @QAWolfHQ
The fix isn't slowing down AI adoption. It's raising the bar on judgment: → Reject code the author can't explain → Make QA adversarial → Deploy cautiously → Teach engineers to interrogate AI output, not just use it Read our piece below:
The Rise of the -10x Engineer: The Negative Side of AI Productivity | QA Wolf
From qawolf.com
38
QA Wolf
@QAWolfHQ
May 26
Replying to @QAWolfHQ
The original 10x engineer was rare, not just because of their output, but because of their judgment. They knew which shortcuts were landmines, what to build, and what to throw away. The AI-assisted "10x engineer" often replicates the output without the judgement.
47
QA Wolf
@QAWolfHQ
May 26
Replying to @QAWolfHQ
The -10x engineer looks like a great hire. They care about making an impact. Their PRs are frequent. Their metrics look great. Right up until they cause a site outage that costs millions. Their REAL impact:
32
QA Wolf
@QAWolfHQ
May 26
Replying to @QAWolfHQ
Here's the math: AI-generated code has 1.7x more bugs than human-written code. That means you're more likely to have -10x engineers on your team than 10x engineers.
The Rise of the -10x Engineer: The Negative Side of AI Productivity | QA Wolf
From qawolf.com
31