Blog
Stack Overflow Business
Stack Internal: the knowledge intelligence layer that powers enterprise AI.
Stack Data Licensing: decades of verified, technical knowledge to boost AI performance and trust.
Stack Ads: engage developers where it matters — in their daily workflow.

Issue 344: Agentic overload

It seems like not so long ago we were tinkering with our chatbots and marveling at how good they were at writing LinkedIn posts. Nowadays, we’re seeing real transformations in the value and creation of code from agentic AI workflows. In this Overflow, we’re looking at what happens to our SDLC when code is cheap, but good judgment, specifications, and QA are still priceless.

On Leaders of Code, we sit down with Google’s Andi Gutmans—co-creator of PHP—to discuss how managing teams of agents is the natural next chapter in software history. In our latest dispatch from O’Reilly, we’re finding out why the "right amount of spec" is the real secret to agentic development. We also chat with Motorola’s Suneet Malhotra on adopting a QA mindset to build reliable agentic SDLC pipelines, and explore why WordPress VIP's Brian Alvey thinks we need to build the future of our internet for both bots and humans.

From around the web, we’ve rounded up a collection of real-world dilemmas, security traps, and digital absurdities, everything from the dangers of fake interviews to your rights as a consumer. And of course, we’re bringing you plenty of knowledge from our Stack users. This week, you might learn a thing or two about dealing with your brain-rotted kid to the ethics of solving math theories using AI. Well, what are you waiting for? We’ve got an overload of links and thinks for you down below.

From the blog

Dispatches from O'Reilly: The right amount of spec for agentic development

When code gets cheap, the hard part is deciding what “correct” means and building a reliable way to check it.

Get rid of your CAPTCHA, the future of the web is bots

Ryan chats with Brian Alvey, CTO at WordPress VIP, about how AI agents are changing the business models of the web, what parts of how we build our sites won’t survive our current digital evolution, and why your site will always need structured content.

From PHP to team lead of agents: rethinking judgment, review, and data with Google's Andi Gutmans (Part 1)

Andi Gutmans, head of Agentic Data Cloud at Google and co-creator of PHP, joins Leaders of Code to talk about why agentic development feels less like a break from the past and more like the next chapter of the same story. This is part one of a two-part conversation.

Building an agentic SDLC with a QA engineering mindset

Ryan welcomes Suneet Malhotra, Senior Manager of Test Engineering at Motorola Solutions, to chat about building end-to-end agentic SDLC pipelines using MCPs, using Cohen’s kappa to evaluate multiple LLMs-as-judges, and how you can improve requirements by shifting QA left through a specification enrichment stage immediately after the design phase.

Raytheon Engineers Build What Others Depend On

At Raytheon, engineers turn complex challenges into advanced technologies that connect and protect. Bring your expertise to teams developing smarter defense capabilities service members and allies depend on.

Interesting questions

Chaos inside my home due to mobile phones

Who would have thought that the brightly lit dopamine machines would become so addictive to children?

Seems like the student’s main advisor fully wrote the paper draft with AI, what to do?

Maybe evaluate the paper instead of asking ChatGPT if it was written by AI.

AI agents as independent legal entities

But what if your dad’s new AI girlfriend is added to his will?

About AI and how we publish

Seems mathematicians are finally doing the calculations on the costs of AI.

Links from around the web

Consumer Rights Wiki—anti-consumer practices database

In case you didn’t know, we do have some rights.

How to compromise your system with a job interview

Tread carefully, even if the hiring manager happens to be a Nigerian prince.

GitHub, autoscaling, and the component substitution fallacy

External incident reports never tell the full story.

Readers can't identify watermarked AI text

People fear their AI slop getting even sloppier.