Title slide saying "One Developer, Two Dozen Agents, Zero Alignment Why we Need Collaborative AI Engineering"

Slide saying "Alignment and communication is the new bottleneck"

Slide saying "Human context is not in the codebase" Business context & financial resources, Political dynamics, Product vision, User research insights, and Organisational history are not in the codebase.

Got to talk at @aidotengineer.bsky.social conf last week about the need for collaborative AI engineering.

All our current coding agents are single player. We're trying to scale up individual productivity, but creating tons of alignment problems in the process.

We have no good tools for...

I see how tight ya'll are running things building Swamp. That's The Way. But I wouldn't want people to think they can let it ride and leave that stuff for later. It's a new game, much less forgiving.

Human-pace development meant people could muddle through without great guardrails (TDD, pairing, proper CD, ...). Yeah, deploying to prod was painful, but everyone jumped in and fiddled with the configs until it was working, and carried on. Can't muddle through like that at agent-speed.

I totally agree with what you're saying. Where I differ: while agents are better at handling bad code than humans, the pace they crank it out means the wheels still come off way sooner than you'd think. So you need to put those guardrails early or it will get away from you fast.

Laziness, Impatience, and Hubris · Adam Jacob How Larry Wall's virtues of a Perl programmer inspire me in this era of engineering

The future is not pumping out slop until our software breaks under its own weight. Instead I think we use AI Agents to define a new SDLC. Larry Wall's virtues of a Perl programmer: laziness, impatience, and hubris compel us to. /cc @bcantrill.bsky.social

www.adamhjk.com/blog/lazines...

They're selling everything but the stock ticker? Surely they could have just sold the company and used the money to start a new one. Sounds like shenanigans.

On the other hand, the "Austrian school" is basically a group of seasoned hackers and builders using simple primitives to create genuinely interesting systems.

Pi is the engine behind OpenClaw, a product that took everyone by surprise, and now every company is scrambling to copy it.

And all too often announced as "Hey, look, I've invented this awesome new thing that nobody else has ever thought of!!"

"Hey look, I built my own wheel!!" has never been easier.

Picture of Marjane Satrapi alongside a quote from her. The quote reads: The world is not divided into countries. The world is not divided between East and West. You are American, I am Iranian, we don't know each other, but we talk together and we understand each other perfectly. The difference between you and your government is much bigger than the difference between you and me. And the difference between me and my government is much bigger than the difference between me and you. And our governments are very much the same... - Marjane Satrapi, Iranian-French graphic novelist

Thinking about this quote from Persepolis creator Marjane Satrapi again.

"Absolute AI Maximalist" Adam Jacob on Building Software That Builds Software Stephen O’Grady sits down with Adam Jacob, CEO and Co-Founder of System Initiative, for a candid conversation about what it actually feels like to build software in the age of AI agents. Adam describe...

i've watched @adamhjk.me's journey with AI and thought he'd be a good person to talk to about its impact on builders, and i was not disappointed.

this was a very fun and interesting chat.
redmonk.com/videos/adam-...

And now he keeps referring to it as "t-shirt window" 😂

Looking for work. Will DevRel for coffee or wine 😁

The biggest leadership downgrade in the history of the world.

There's a nice #Artemis II mission tracker here: artemis.cdnspace.ca

LOTS of data about its current position, crew activities, and whether or not the toilet is currently "Go"

Agreed they are reducing what's covered by plans to reduce losses (and manage capacity). My point was my experience is OpenClaw doesn't use anything like as much as coding. My estimate is a few hundred $/mo, and I'm pretty heavy

I'd guess pushing people onto CoWork is a hope, too.

Christopher Penn wrote: Just remember that given the abundance of neurodivergent people in science, it's far more likely that autism causes vaccines.

A different perspective.
Always helpful.

The Anthropic subscriptions are throttled, and stop working when limits are hit every 5 hours, week, or month. So yes they are cheaper than paying per token but nowhere like thousands

Title: Easter Egg Hunt techniques inspired by great detectives Panel 1: Sherlock Holmes. Deduce the exact location of every egg from seemingly unrelated details without leaving your room. Send your amazed sidekick to fetch them. Panel 2: Jack Reacher Hit the garden hard and fast. Take no prisoners Get back on the road. This approach results in a thrillingly intense hunt, but very few intact eggs. Panel 3: George Smiley Meticulously assess documents, people and motives. Then scrupulously collect the eggs and dejectedly wonder if any of it was worthwhile.

My Easter books cartoon for @theguardian.com

Awesome photo! But contrary to some commenters I've seen on here; no, you don't see the ISS, or satellites. And no, LEO debris doesn't create a perceptible haze.

I heard them on the livestream yesterday telling mission control that their GoPros and iPhones weren't taking good photos

Harness engineering for coding agent users A mental model for building trust in coding agents through feedforward guides, feedback sensors, and iterative harness engineering.

New article where I offer definitions & a mental model how to think about harness engineering as coding agent users. Building blocks at our disposal, dimensions and goals to consider; emerging practices, open questions; and of course, what role do humans play martinfowler.com/articles/har...

We can try to find utilitarian rationalisations, but we're doing it because curiosity is an essential human trait. We're going to the moon because it's there, and we can.

They're going further out than any previous mission, so maybe they can get a shot with both moon and earth fully in the shot? That would be something!

Harness engineering for coding agent users A mental model for building trust in coding agents through feedforward guides, feedback sensors, and iterative harness engineering.

NEW POST

@birgitta410.bsky.social wrote some initial thoughts about Harness Engineering last month. Since then she's been researching more and has now written a thoughtful mental model for understanding the topic.

martinfowler.com/articles/har...

As we build, so we believe · Adam Jacob Building the machine that builds the machine, in order to learn what I believe about the future of my profession

I know that for me, the only way I'm going to find out what I believe is by actually building things on that frontier. If you're feeling anxious about it, or uncertain - might I suggest you do the same?

www.adamhjk.com/blog/as-we-b...

People in the sixties: I better not say that or the government will wiretap my house People today: Hey wiretap, do you have a recipe for pancakes

This

"But what was the ROI?"

This post by @doctorow.pluralistic.net on corporate bullshit receptivity (pluralistic.net/2026/03/19/j...) has echos of @vgr.bsky.social's Gervais Principle (www.ribbonfarm.com/2009/10/07/t...).

Sociopaths have high Organizational Bullshit Perception Scores, while the Clueless buy into it.

The "are we the baddies" meme, but with an orc

Apparently people who work at Palantir call themselves Hobbits?