Products
Stack Overflow for Teams
Capture, share, & collaborate on knowledge internally.
Advertising
Promote your product or service to developers and technologists.
Talent
Engage the world’s technology talent with your employer brand.
From the Network
biology.stackexchange.com

Do trees have epigenetic clocks?

If a tree ages in a forest with no one around to see it, did it really age at all?

english.stackexchange.com

What is this obfuscation method called?

Your secret language shares something in common with teens trying to talk in code online—It's all just Leet.

law.stackexchange.com

Can one debt cancel another debt?

The law does not have an official way to say, "I got you this time, you get me next time."

academia.stackexchange.com

Frustration with students and deadlines

"No." is a complete sentence.

Apple Podcasts Overcast Pocket Casts Spotify RSS feed
Releases
Latest articles
More Podcast
Around the web
tonsky.me

Needy programs

When did using the internet start to feel like dealing with a clingy ex?

spectrum.ieee.org

A challenge to roboticists: my humanoid Olympics

How many robots does it take to turn a sock inside out? A lot, apparently.

arpitbhayani.me

Heartbeats in distributed systems

Hopefully the constant beating of all the nodes in your distributed system won't turn you into an Edgar Allen Poe character.

raganwald.com

A brutal look at balanced parentheses, computing machines, and pushdown automata

Taking "software architect" to its most literal interpretation.

aifoc.us

Dead framework theory

Every LLMs' favorite framework is apparently React.

prison.josh.mn

Startup lessons from my piracy website

Well, make sure your antivirus is up to date before you start going on PirateBay and LimeWire for business advice.

codemanship.wordpress.com

Is software the UFOlogy of engineering disciplines?

The truth is out there…but so are lies. And bugs.

mccd.space

Why I love OCaml

Could the French by the proprietors of taste even in programming languages?

anthropic.com

Emergent introspective awareness in large language models

Officially, you can be less emotionally available than a Claude bot.

phys.org

Mathematical proof debunks the idea that the universe is a computer simulation

This has to be the workings of Agent Smith.

plzoo.andrej.com

The programming languages zoo

Lions and tigers and Lambda, oh my!

vanderessen.com

The human only public license

The last refuge of humanity lies in license agreements.

Want updates to your inbox?

Every week we’ll share a collection of great questions from our community, news and articles from our blog, and awesome links from around the web.

or edit your settings on your profile page.

Issue 303: To abstract or not to abstract

To abstract or not to abstract? That's the question on our minds this week. Whether you're looking for more transparency or just trying to get your code out the door, we have plenty of stories for you. On the pod, we've got a chat with Graphite's Greg Foster on how to make your AI-generated code more secure. Spoiler: it's by giving humans more context and visibility into the code. We also spoke to Nic Benders from New Relic on the complexity crisis, and how developers don't just need observability, they need understandability. And if your feelings about abstraction and AI aren't complicated enough, make sure you read our blog on how AI is abstracting human creativity, written by our very own Eira May. All this talk about hidden layers has us wanting to be more transparent, too. We're sharing what's new on Stack Overflow for November—from voting, to anti-spam, to open-ended questions—in our first ever monthly update blog. Being a software developer can sometimes make you feel like you're in The X-Files, which is why one developer from the web is calling software the UFOlogy of engineering, and another is worrying about the death of frameworks because of LLMs' abstraction. But if you're ever worried about too much opacity, you can always go old-school and change your programming language; we have at least one dev in this week's issue who really thinks you should use OCaml if you do. But one thing hasn't changed: if you don't ask, you'll never know. So we have plenty of questions and answers this week that should give you a little clarity into the abstract. For instance: if something happens and I don't see it, did it really happen? Is a bot's Spotify Wrapped less valid than mine? Are the lyrics, "What if God was one of us?" still applicable if God has no birthday? Unabstract those abstracts in the links below.

Issue 302: Get your specs right

Developers get to use a lot of tools that both hurt and help their productivity. But over here at Stack Overflow, we think AI agents might be some of the best and brightest of them all. On the blog, Ryan talks about how agents are poised to become the best tool for developers, because in his words: One tool is better than ten. On the pod, we welcome back Deepak Singh from AWS to talk Kiro, the spec-driven coding agent his team is building, and how he imagines the work of developers evolving. But don't let the tools get you too excited about kicking your feet up at your dev job—Tom Moor from Linear joined us to discuss the importance of context in agent effectiveness, and how junior devs can shape up for this new era of tech. Speaking of junior devs, our CEO Prashanth Chandrasekar has an upcoming convo with OpenAI about learning to code in the age of AI, and we have the details on that one for you on the blog. Whether you're a junior dev or one with years of experience, you could always build out your skillset a little, so we've got a whole zoo of mini programming languages to help you polish up your understanding. We've also got stories on introspective bots, and how to keep your own introspectiveness away from them through a humans-only public license. If you're feeling stuck in the Matrix right now, don't worry—be sure to check out how a mathematical proof debunks any fear you might have about being stuck in a simulation. But could that mathematical proof just be part of the Matrix anyway? One user on our math site is wondering what mathematical proofs were developed using LLMs We've got that answer for you and more—about backwards wheels, loud electric cars, and Gandalf's amnesia—ready for you in the links below.

Issue 301: The agents are here.

If you're keeping your ear to the beat of AI's robotic heart like we are, you're probably noticing that agents are everywhere now. This week's newsletter is not different—we've got plenty of stories for you on the newest AI hype. John Dickerson, CEO of Mozilla.ai, joins us on the pod to talk about the role of open source in the future of AI agents, including the challenges OS communities are facing in the age of AI. If you're looking to throw away your runbook, we've got another conversation with Spiros Xanthos, CEO and founder of Resolve AI, all about how AI agents are reducing the work of incident management and troubleshooting for developers. Plus, on the blog, find out what it takes to build out agentic virtual assistants on top of decades of ebike manuals." If you're saying, "Enough about agents, what else is going on?", don't worry, we have plenty of other stories for you. On Leaders of Code, our very own Natalie Rotnov dives into the 2025 Developer Survey and what the results mean for tech leaders. From the web, we've just learned that LLMs—much like your teenager—can get brain rot. But don't worry, if your AI starts spamming memes you don't understand, you can just go back to working on a 1980s ZX Spectrum. We have that story from the internet, plus a story about building software for a different kind of rot, ready for you in this issue. And just to prove that agents aren't the only things that people are talking about, we have plenty of questions and answers for you. Wondering what is and isn't an indie game? Mostly, it's the vibes. How do you extend your WiFi through an old house? We hate to tell you this, but you probably need to use an Ethernet cable. How do you keep your kids from biking straight into oncoming traffic? Like any parent, you just have to try your best. We have those answers and more in the links below.

Issue 300: This...is... STACK OVERFLOW!

It's Issue #300 and we're keeping it Stack-Overflow-classic by answering all your questions. How can AI agents make everyday life easier? We got that answer for you on the pod when we chatted with Yutori's Dhruv Batra about building proactive AI agents. How do you achieve digital sovereignty? The answer: start by building your own cloud. Check out our conversation with Andrei Kvapil from Cozystack to hear how open-source has made that possible. Is someone going to attack my JS code? Probably, but we have a blog on ten ways you can keep your precious frontend secure. But that's not all. We've got the story on how a pesky dot in Turkish broke Kotlin, and the tale of a dev who's creating one million nodes in a single Kubernetes cluster. Elsewhere on the web, you won't believe this one weird trick for creating a beautiful UI (spoiler: it's a spreadsheet). Plus, if you've ever said, "You just don't get it," when arguing with someone about AI, we have a piece for you. But would we be Stack Overflow if we didn't bring you the answers to questions you never thought to ask? Should you feel bad every time you misspell "accomodate" or is it actually the dictionary's problem? Is the shady letter from your bank legit and should you send them Bitcoin to make it right? Is it a threat when your mom says she "loves you to bits"? Just like #1 through #299, Issue #300 is overflowing (get it?) with answers for you down in the links below.

Previous issues