Can our billionaires in America develop a sudden interest in architecture and start building some grand things in our cities? We may as well get cool buildings and environments in exchange for our dystopia

I didn’t have any sunglasses, so I worked with what I had

“You can’t defend capital and people at the same time” … helluva quote

I did it, I wrote it. The thing I think (?) some educators have been waiting for for quite a while. mikecaulfield.substack.com/p/sift-for-a...

My perhaps controversial assessment is that when you share the propaganda of your enemies to call it out or to mock it… you’re still spreading their propaganda and maybe doing exactly what they hope you’ll do.

Things are heating up in the AI wars! #Sora sora.chatgpt.com/p/s_68e01655...

Anyone got a Sora invite you’re dying to give away? 😅

I think something democracy has never faced before is a populace with years and years of algorithmically supercharged perceptions of “the other side”

To be honest though ... ChatGPT's sycophancy is helping democratize access to that multi-layered-bubble of agreeable yes-men 😅

Ever worked with a "Cowboy Coder"? Or perhaps, you yourself are that gun-slingin' cowpoke! What works in a smaller team or startup, can eventually become a liability as a team grows if not managed correctly. codecube.net/2025/9/team-...

A recent tweet by my friend got me thinking about the many winding paths I’ve been down over the course of my career … starting up a new series on my blog to write about it :) codecube.net/2025/9/accid...

Yep. Almost all of history from global warming, mass migration, 9/11 and the rise of Russia is downstream of not aggressively pursuing nuclear power. When you look at the body count of misinformation, misinformation about nuclear power ranks near the top.

Does your team "move fast and break things"? Or do you slow down and find/fix the root cause every time? As always, the right answer is the classic "it depends", and each approach comes with its own pros and cons. codecube.net/2025/8/team-...

As coding agents get more capable, it seems like they’re using more and more compute to handle longer and more complex tasks. Devs will increasingly have to start rationing where they do their AI thinking to avoid burning through their credits. codecube.net/2025/8/tiers...

Finally got this working ... this had been on my personal backlog for quite some time, and so now I can embed timelines in my blog posts 📆 codecube.net/2025/8/timel...

I got a Mountain Dew Baja Midnight, out of curiosity… I’ve made a huge mistake

Has "Agile" ever actually worked for anyone? 😅 new blog post where I'm just kind of retrospecting across the last 25 years codecube.net/2025/8/agile...

So ... is this the reference song that Soda Pop was based on in #KPOPDEMONHUNTERS? open.spotify.com/track/1mWdTe...

Had a fun time working on a new github action for @wildernesslabs.bsky.social's Chloroplast (a static site generator for writing and publishing docs). Did a quick writeup on it in my blog: codecube.net/2025/8/chlor...

People on social media should remember that propaganda doesn’t always need to be political in nature … (try to) know when you’re being manipulated

How many engineers have heard their leadership say, "why do things take so long to ship these days?" ... there's probably a bunch of answers to that, not the least of which is that the things that work for a small team, cause pain for larger teams. codecube.net/2025/8/team-...

I got into tech because it’s fun to take things apart, see how they work, and put them back together — or rearrange the parts into something brand new. it definitely feels like the new ethos is “bro why bother to learn? caring about that stuff is cringe, just go FAST bro” and that bums me out

Dang it, debugging CI code is such a gigantic pain ... in this case, a custom github action to easily run #Chloroplast documentation builds. It's gonna be so cool once it works!

Before I get banned or blocked for going about this the wrong way… Does anyone know the best way to pull things out of the Internet way back machine? I’d like to go in and extract some of my very early blog posts from the early 2000s that I don’t have.

GPT 4o after people started checking out GPT5

What movie have you seen an unhealthy number of times?

On the eve of GPT5 launching, I start to wonder how the future will evolve. As we keep integrating these new tools into our workflows, will we (collectively) get to keep the generated value, or will it be increasingly extracted from us. codecube.net/2025/8/ai-br...

just a reminder that sustained overwork leads to exhaustion and increased error rates — sustained 60 hr/wk employees are actually LESS productive than if they worked 40 hours

Stop it, tech industry 🗞️ www.forbes.com/sites/bryanr...

Nice update just dropped for @wildernesslabs.bsky.social's Chloroplast ... it now generates sitemap XML files for an easy boost to your documentation site's SEO 👏🏼 Available via Nuget! github.com/WildernessLa...

Feels like with every passing day, individuals are asked to take on a wider array of responsibilities; in previous years they might have been the responsibility of someone who specialized in that role. Increasingly frayed engineers ask, is this the only way forward? codecube.net/2025/8/team-...

Doing a bit of experimenting with some alternate navigation mechanisms on my website. Does anyone have any good examples of graph based in navigation on different websites? codecube.net/2025/7/link-...

I think a lot about the “harmless” part of “helpful, honest, harmless” which has been the (vague) rallying cry of AI Alignment. I consider AI doing student homework to be harmful at the individual and societal levels. Very much in conflict with “helpful”.

Why is everyone talking about clams?

Another weekend project launched. This time for the MCP community! VS Code - MCP Install Button Generator Check it out at VSCodeMCP.com The backstory: I wanted an easy way for users to one-click install my Lokka.dev MCP Server to VSCode

Possibly my most worthless post ever 🙈 The world is changing, and those of us that "exist on the internet" should understand that what is valued is changing before our very eyes (for better or worse) codecube.net/2025/7/worth...

Saw Superman … it was pretty wonderful 🐒 #SuperShit

This has probably been the biggest issue with the community growing here on bluesky ... everyone's obsessed with still 👀 what's going on over there. And the more everyone does it, the more folks here get a weird sense of FOMO. Leave that mess there! 😅

Have you ever felt like every day was just a procession of one meeting after another? Eventually you ask yourself when exactly it is that you're supposed to do any of the work that you're talking about?

It’s always fun going back and looking at old posts … I was poking around my sites archives, and came across a post from 2013… Where I was just kind of reflecting on how work was changing. Wrote a post, reflecting on how things have gone since then… codecube.net/2025/7/how-w...

I know this probably is beside the point as far as time savings go ... but I kind of enjoy watching the github copilot session and watching it reason through the task 😅

You know, I’ll say it… I miss the community that built up around newsgroups in the early 2ks, along with the early “blogosphere”. The move to crowdsourced sites like slashdot, digg, hackernews, and eventually StackOverflow and other social media diluted and degraded the online community.

So I've been using Github Copilot directly on github recently ... and generally speaking it's been great! Honestly, sometimes when I'm not in a huge rush, I weirdly enjoy just watching the live session as it reasons its way through the task I've assigned to it. 🧵

Having the need and opportunity for your engineering team to grow can be exciting ... but there are potential pitfalls along the way: codecube.net/2025/7/team-...

xAI’s competitive advantage was supposedly that Grok could be trained on proprietary access to tweets on X. Now that seems like competitive disadvantage. 😬

The timing of this response to a 3+ year old tweet of mine is just impeccable

Finally getting around to writing a series I've wanted to write about for some time, talking about how teams grow and all of the potential failure modes therein. I'll keep posting about it as I write new entries, but here's the link to the first post in the series: codecube.net/2025/7/team-...