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 329: The heart of our technology

This week on the blog, we’re diving into all the things that make the tech we know and love possible—architecture, hardware, and databases. We sat down with AMD’s CTO Mark Papermaster for a conversation on silicon and chips, and how AI is helping solve the chip shortage it caused. We’re also pulling back the curtain on our own architecture and how we’re replacing Ingress-NGINX at Stack. Plus, we’re joined by Qdrant’s Brian O’Grady to decipher what exactly we mean when we talk semantic vs. vector search.

Beyond the blog, we’ve got a hearty round-up of everything you’re curious about, from phishing attempts to robodogs to piston engines. And there are no dumb questions here—kind of like our new Friday column on MCP servers. Are we all pronouncing “the” incorrectly? Is human imprecision the heart of software development? Is it so wrong to fall in love with a bot? Is Cloudflare trying to hack you? We’ve got all those answers and so much more to satiate your curiousity down below.

From the blog

What (un)exactly do you mean by semantic search?

Ryan welcomes Brian O’Grady,  Head of Field Research and Solutions Architecture at Qdrant, to discuss the differences between traditional text search engines powered by Lucene and modern vector databases, when vector search’s exact-match needs work for things like logs and security analytics and when semantic search works for user-facing discovery and non-exact results, and how Qdrant is growing into video embeddings and local-agent contexts.

How we replaced Ingress-NGINX at Stack Overflow

Ingress-NGINX had been handling our traffic routing since moving to Kubernetes, but when it was announced it would be retired, we were forced to consider a new traffic routing solution.

AI giveth and AI taketh CPU

Recorded on the floor of HumanX, Ryan is joined by AMD CTO Mark Papermaster to discuss AMD’s silicon strategy for AI borne of their long history of heterogeneous CPU/GPU computing, how chipmakers are dealing the wide range of AI workloads from training to inference, and the paradox of agents both eating up all the compute and helping AMD accelerate chip innovation.

No Dumb Questions: What is an MCP server and why do I care?

Welcome to No Dumb Questions, a column where our least technical writer asks our technical staff the simple, basic tech questions people are afraid to ask. In this first entry, Stack's Director of Ecosystem Strategy Ben Marconi teaches us the basics of MCP servers and why they matter.

Interesting questions

In a piston engine, must “cylinders” be round?

Anything can be a cylinder if you just believe.

The + vowel letter

At this point, English’s imprecision is what makes it English!

Possible red flags with this recruiter's behaviour

Who knew the best way to get a job at a company was to apply directly to the company for the job?

Cloudflare is asking me to run a command in PowerShell on visiting my friend's site

Probably better to ask about the suspicious action before you perform the suspicious action, but that’s just me.

Links from around the web

The bottleneck was never the code

AI haters despise this one simple trick—precise Roadmaps.

CARA 2.0

The robotics race has come down to who can build the goodest boy the fastest.

MIT engineers’ virtual violin produces realistic sounds

Now you can make a lofi bedroom beats version of Vivaldi’s The Four Seasons.

Three inverse laws of AI

I’d add, “Don’t fall in love with it, even if its voice sounds like ScarJo.”

Onboard, organize, and bring your team up to speed in a jiffy. Try Stack Internal.