This week’s Linky has a bit of everything. From how AI “explains” itself (and why that’s not the full story) to what it means to be an expert generalist. There’s also something on improving what’s in front of you, chasing your way of working rather than titles, and why you should learn like an athlete.

And if you like these kinds of links and reflections, subscribe and you’ll get them straight to your inbox every week no hunting them down.

Subscribe now

Latest post from the Quality Engineering Newsletter

I shared my takeaways from the BCS Software Testing Conference. It was a great reminder that one of the best ways to learn is seeing what others are doing. If you want more than the highlights, my full notes are here.

I also resurfaced a post from 2023. It looks at who should build a culture of quality. It’s not the only way of seeing it, but it does show how different roles can support each other based on their sphere of influence. This builds on Quality Engineering as philosophy, a framework and a tool.

Conferences

Conference season starts in September. I’ll be at:

• Lean Agile Scotland (Edinburgh)

• TestBash (Brighton)

• Oredev (Malmo, November)

If you’re going to any of them, come say hi and if you're nice I'll give you a QEN discount code. I’ve also got some discount codes for the confs too - just reply to this email or DM me.

LLMs don’t always tell you how they got there

We often ask LLMs to explain their answers using “chain of thought”. This study suggests those explanations might not actually match how they got there. They can silently correct mistakes or take shortcuts without telling you, while still giving you something that sounds convincing.

A good reminder that you can’t always take LLM “reasoning” at face value. Via Chain of thought is not explainability

Did you improve things today?

It doesn’t matter what position you find yourself in right now. What matters is whether you improve your position today. Every ordinary moment is an opportunity to make the future easier or harder.

It’s a nice way to think about making complex systems healthier. Can you leave things a little better than you found them? Just make sure those improvements connect to something bigger, otherwise you end up going in circles.
Via The Three Lenses of Opportunity Cost | fs.blog

Quality engineers as expert generalists

We've seen our best business analysts gain deep skills in a couple of domains, but use their generalist skills to rapidly understand and contribute in new domains. Developers and User Experience folks often step outside “their lanes” to contribute widely in getting work done. We've seen this capability be an essential quality in our best colleagues, to the degree that its importance is something we've taken for granted.

ThoughtWorks are now looking to cultivate this quality by calling them Expert Generalist. But why did they use the word “expert”?

There are two sides to real expertise. The first is the familiar depth: a detailed command of one domain's inner workings. The second, crucial in our fast-moving field is the ability to learn quickly, spot the fundamentals that run beneath shifting tools and trends, and apply them wherever we land.

This is what a quality engineer is to me. We need to know how quality is created, maintained, and lost in our context, and then apply that to people, processes, and products. That means developing our knowledge of behavioural science, agile delivery, and software engineering too. Via Expert Generalists | Martin Fowler

Whose job is safe from AI?

Autor and Thompson’s framework does suggest a clarifying question: does AI look like it is going to do the most highly skilled part of your job or the low-skill rump that you’ve not been able to get rid of? The answer to that question may help to predict whether your job is about to get more fun or more annoying — and whether your salary is likely to rise, or fall as your expert work is devalued like the expert work of the Luddites.

I think the answer to this question depends on your perspective. Testing, for instance, might be considered "low skilled" work for some while "highly skilled" to others. So depending on who you ask some may say that Testing will be taken over by AI and others will say that Testing still needs human judgement.

I think Testers skills sets are going to shift from doing the testing towards requesting what testing the AI systems should do and interpreting the results of that testing for engineering teams. They'll retain some forms that they enjoy for themselves but outsource what they don't or can't do to AI.

This feels closer to quality engineering: using your knowledge of how quality is created, maintained and lost to guide what’s tested, then leaving the execution to machines. Via Whose job is safe from AI? | Tim Harford

Chase your way of working, not a title

Swap “lifestyle” for “way of working”...

"Chase your desired lifestyle, not your desired title. People are blinded by status and labels. Once you release the need for a specific title, there is almost always an easier path to living your preferred lifestyle."

Instead of waiting until you have “quality engineer” in your title, work like one where you are now. You’ll still have boundaries, but you might be able to bend or add things in ways that get you closer to your desired way of working. Sometimes you might even find teams who want you to work like that too. Via 3-2-1: On avoiding frustration, easier paths to a good life, and the power of not reading | James Clear

Train like an athlete

"Athletes train. Musicians train. Performers train. But knowledge workers don’t."

Top performers train deliberately. They're clear about what they want to get better at, and they keep at it.

I've got a few ways I "train": a backlog of articles to read, 2-3 books I dip in and out of, I write a long-form post every two weeks, usually have conference prep on the go (which usual involves lots of research), and plenty of note-taking from all the things I'm consuming. I walk a lot too - that's when I do most of my thinking and try and make sense of things. It's not perfect, but it keeps me learning deliberately rather than drifting.

I suggest that you don't try to match other people's learning regimes but use them as inspiration. Each of us has different responsibilities and things we need to do. You need to find what works for you. Remember, a little is better than nothing, and often it's taken people years to develop the routines they have. Start small and iterate, just like how we build quality in. Via Learn Like an Athlete - David Perell

If you’re new here and enjoyed this, consider subscribing. It’s free, and you’ll get future Linkys and posts on quality engineering, learning, and working better together.

Past Linky Editions