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 343: There has to be a better way

Ever since the dawn of technology, we’ve been trying to figure out better ways to do things. Whether it’s replacing pens with the printing press or adding layers of abstraction to everything we do, we’ve always used technology to make our lives easier. Still, there’s bound to be some missteps when you’re evolving so fast, and this Overflow is dedicated to correcting those mistakes in the latest wave of AI hype. Here's how you get on the good foot and out of the hype: stop “tokenmaxxing”, run hackathons instead of stuffy integration projects, and figure out you agent's context architecture.

And we’ve got stories past just the blog and pod about AI missteps this week. For instance, are the big guys in the industry making a huge mistake with how they handle AI watermarking? Has AI slop infiltrated the highest orders of academia? How brainrotted will our kids get if we teach them about AI before we teach them about real code? Find you footing and head over for all that and more in the links below.

From the blog

No Dumb Questions: What is AI context architecture? Why not just build your own?

In this No Dumb Questions, Phoebe asks Stack’s Engineering Manager Doug Whitley and Product Manager Ash Zade everything she wants to know about AI context architecture. What exactly is it? Why is it so important? What makes for good AI context architecture? Why buy one when you can build your own?

Solving integration woes with a hackathon

Ryan welcomes Meryll Blanchet,  Director of Engineering for Adobe Brand Visibility, to chat about Adobe’s recent acquisition of Semrush, how Adobe Brand Visibility was born from Semrush’s AI visibility product and Adobe’s LLM Optimizer, and how Adobe used a three-day internal hackathon instead of a large-scale infrastructure integration to quickly deliver value to customers.

Your tokenmaxxing is not valuemaxxing

Ryan is joined by Coder’s Rob Whiteley to chat about why tokenmaxxing isn’t proving real value and just triggering Goodhart’s Law, how release speed and PR merges can help you measure agentic outcomes with or without a human-in-the-loop, and what the democratization of skills means for junior developers and the talent pipeline.

Interesting questions

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.

As a tutor, I have a 7th grader who wants to learn Python, and her parents want it tied to AI

Make sure every time you say "AI" you do air quote fingers.

Teaching newcomers: where is the line between meeting someone where they are and validating their wrong views?

Forget everything you knew about Buddhism. That's the first step to understanding Buddhism.

What makes us perceive something as real?

Asking for an imaginary friend.

Links from around the web

How I developed an Am29000 C compiler and web browser

And it’s all backed up on floppy discs, like the software gods intended.

Taking a tip from plants—for the eyes?

Now you can use plants to fix the dry, red eyes you get from using a different leafy green weed…

Meet the only known trebuchet casualty in history

“Skeleton 150 was in terrible shape, even for a dead guy.”

How AI text watermarking works

So they don’t just put, “Written by Claude,” in invisible ink at the end?

Spending hours searching for answers at work? Find them faster in Stack Internal. Get it free!