In The Observatory, the MoTaverse community shows up to share links they are reading all the time.

The value is not only in the links shared in the moment, but also in the accumulation of links that become searchable knowledge over time.

There are currently 3126 searchable links.

Hello Security World

Join to launch the chapter and shape future events.

🎉 Featured link of the week:

Don’t talk to me about System Integration Testing — Ash Winter

I have a real thing about “System Integration Testing.” Not the type of testing, thats neccesary sometimes. Mainly how excessive amounts of it result from wicked organisational problems, you know the ones. Horrible, cultural bottlenecks that no amount of AI will save us from, until it replaces software and organisations entirely I guess. Although its having a good go at killing off software-as-a-service, which is responsible for a lot of system integration horror stories.

Top links from the Observatory this week:

Accessibility and the new model of the internet. — Cassidy Williams

I’m really grateful to work at a place like GitHub where accessiblity is really taken seriously.

But, industry-wide, I feel like we haven’t seen as much care and dedication to accessibility historically.

38 things we've learned about building AI products — Ian Vanagas

We’ve built 10 different AI products at PostHog, including PostHog AI, our install Wizard, Replay Vision, Scouts, and more. They’re used by nearly 100,000 people every month. This is what we’ve learned.

The world is not a beta test — Gaping Void

A few years ago, some members of our team sat in on a very interesting panel at a university conference titled Should Universities Be Run More Like Startups?

On the face of it, it sounds like a great idea. Faster innovation, more risk-taking, less bureaucracy – what’s not to like?

Why Almost Nothing You Read Online About AI Is True - Codemanship
I’ve adapted my Special Theory of Autonomous Agent Reliability (STAAR) to predict the reliability of social media* posts about “AI”.

🦾Quality Robots: Would you like a Robot, Duck, or a Runner? — Judy Mosley

My LinkedIn feed is full of AI, whether it’s people writing about AI or using it, or attempting to untangle themselves from it. AI is talked about in my friendships, on my favorite podcasts, and in my social media feeds. The discourse around AI, in varying degrees, is everywhere. It’s easy to think we all know about AI and how it’s used in every organization.

Turns out, that’s not true.

You can just ask AI to do things — Sean Goedecke

One of the most important skills in working with AI is agency: being willing to try new and ambitious things. This is true for what you build, but it’s also true for how you build. AI models are improving so quickly that most of us are bottlenecked1 on noticing that AI agents can help automate particular tasks. Here are some examples of that from my daily work.

The software industry has become obsessed with automating testing — Melissa Fisher

Look at almost any test engineering job description and you’ll find the same expectations: build automated frameworks, write automated tests, integrate them into the CI/CD pipeline, increase coverage, and now, increasingly use AI to generate tests. The tools change, the technologies evolve, and the expectations grow. But the underlying assumption remains remarkably consistent: if we automate enough of our testing, we’ll get better software.

From Manual QA to SDET: How to Make the Move — Julia Pottinger

The job listings that used to read “Manual QA” or “QA Analyst” read “SDET” now, short for software development engineer in test, usually with a line about AI tooling tucked into the requirements. The title moved toward engineering. The work underneath it moved less than the title makes it look.

Four things quality engineers can learn from recent AI exploits — Jesper Ottosen

Sometimes it’s up to the quality engineers, QA and testers to do all kinds of application security testing – especially in smaller firms, where the role boundaries aren’t that specific. In bigger companies it’s more often seen that it’s up to the security people to do security – and it is a whole world on it’s own. Yet being quality advocates we must also involve ourselves in security testing.

Agentic Verification — Addy Osmani

If you give an AI agent a way to independently verify its work, that can heavily improve the quality of output. Agents write code “much better on average at scale”. The “at scale” part comes from verification. Agents make checking cheap.

Explore all Observatory links…