Building tools for humans.
London
Joined April 2009
- Tweets967
- Following866
- Followers486
- Likes5.5K
In the age of Agentic Engineering
The two most important questions are now:
1. What do people want?
2. How do they find out about it?
Simple questions, but not easy to answer
Few
Great post. Something I've been thinking about and trying to articulate.
Eli Goldratt's book, The Goal, was famous for its (then unpopular argument) that keeping every machine running 24 hours a day, the metric most plant managers cared about, was actively making factories worse. I suspect we're seeing the same fallacy in how many people are using AI agents.
Goldratt's point was that machine utilization isn't throughput. What you want from a manufacturing plants is making good widgets as cost-effectively as possible.
It doesn't necessarily follow that running your machines all the times optimizes that.
Picture a three-station assembly line. Stations 1 and 2 each crank out 200 widgets an hour. Station 3 can only handle 100. Running stations 1 and 2 around the clock doesn't ship more product. It just piles up half-finished widgets in front of station 3, ties up cash in inventory, and creates more work managing the pile.
He developed the Theory of Constraints to point out that what matters is solving the bottleneck in the system, not increasing machine utilization.
I suspect a lot of agent usage right now is the same fallacy at higher resolution. Running 20 Claude Code sessions in parallel can feel productive because something is always happening. But, if the bottleneck in your work is judgment about what's worth doing, more agents just generate more output for you to wade through.
This is not to say there aren't workflows running 20 agents in parallel very effectively, I'm sure there are. And, I suspect there's a general retraining we all need to do around evolving historical workflows. But....
The constraint for most knowledge work is deciding what's worth executing and no one is task switching between 20 things at the same time effectively I don't think. I find I can run maybe 2 or 3 things in parallel with maybe 1 or 2 admin-y type things on the side and that is only if I'm very locked in.
Claude 4.6 is a good programmer but writes insanely severe bugs constantly, it won't catch them all in audits, nor will other claudes
You need codex 5.4 auditing every commit 4+ times. If you don't believe me, try it.
I have an /auditcodex skill for it
github.com/sterlingcrispin/c…
Once you understand this, you can't unsee it.
Agency > Intelligence.
A simple concept that has huge implications.
Kudos to @george__mack who saw it early and dove deep on it: highagency.com/
Agency > Intelligence
I had this intuitively wrong for decades, I think due to a pervasive cultural veneration of intelligence, various entertainment/media, obsession with IQ etc. Agency is significantly more powerful and significantly more scarce. Are you hiring for agency? Are we educating for agency? Are you acting as if you had 10X agency?
Grok explanation is ~close:
“Agency, as a personality trait, refers to an individual's capacity to take initiative, make decisions, and exert control over their actions and environment. It’s about being proactive rather than reactive—someone with high agency doesn’t just let life happen to them; they shape it. Think of it as a blend of self-efficacy, determination, and a sense of ownership over one’s path.
People with strong agency tend to set goals and pursue them with confidence, even in the face of obstacles. They’re the type to say, “I’ll figure it out,” and then actually do it. On the flip side, someone low in agency might feel more like a passenger in their own life, waiting for external forces—like luck, other people, or circumstances—to dictate what happens next.
It’s not quite the same as assertiveness or ambition, though it can overlap. Agency is quieter, more internal—it’s the belief that you *can* act, paired with the will to follow through. Psychologists often tie it to concepts like locus of control: high-agency folks lean toward an internal locus, feeling they steer their fate, while low-agency folks might lean external, seeing life as something that happens *to* them.”
Paul Haydock retweeted
Morning!
After 4 months of work, lots of testing, some tears, I just launched LangSesh! #langtwt
Pic 1: Me, just now, launching.
Pic 2: My 8-day streak on LangSesh from practising Spanish, Esperanto and German.
Pic 3: A voucher you can redeem right now to get 10 free credits.
Paul Haydock retweeted
The greatest enemy of truth is ego. You cannot see through the glass if you're focused on your reflection in the glass. Get out of your own way, unshackle your identity from your views, define yourself only by a willingness to see, and you might glimpse a new and deeper horizon.
Paul Haydock retweeted
People often get things backwards when trying to improve their lives.
They think the path to happiness is: get more stuff, then do what you want, then be happy.
But that's not how it works. The real sequence is the opposite.
First, figure out who you really are. Then do the things that naturally follow from that. Finally, you'll end up with what you truly want.
Just finished @shreyas' Product Sense course on Maven.
Easily one of the best courses I’ve ever taken — maybe *the* best.
Packed with wisdom, not just on Product Sense, but for life too.
Intense, fun, and full of humour.
💯 recommended!
My new favourite Design term "Cognitive Perplexity"
xitter.cf/i/grok/share/8zlow4C16…
Hacking on ollama this weekend. Everything about it screams minimalism.
> Number of key/mouse strokes from install to CLI chat to python API cannot be smaller.
> "Cognitive perplexity" is very low. Our brains naturally predict what should be the expected action when using a new interface. If you think of that as "action token", then the less surprise we get, the lower perplexity it is.
> Not a single button on the website is wasted. No distracting pixels that take up visual processing bandwidth.
Developer UX for advanced AI is a criminally underrated topic. Every library should strive to minimize cognitive perplexity.
Thanks for making my Saturday a joy.
“Everything that needs to be said has already been said.
But since no one was listening, everything must be said again.”
- Nobel Prize winner, André Gide
Paul Haydock retweeted
You will learn more about how your product is used by just talking to 10 users than almost any other method out there. Ask them what they do, what they’d don’t like, and what they’d want to do. It takes an hour, and you’ve probably learned as much as a month of analysis.
..."a calculation that takes Willow under 5 minutes would take the fastest supercomputer 10 to the 25 years.
That's a 1 with 25 zeros following it, or a timescale way longer than the age of the universe."
😲
Introducing Willow, our new state-of-the-art quantum computing chip with a breakthrough that can reduce errors exponentially as we scale up using more qubits, cracking a 30-year challenge in the field. In benchmark tests, Willow solved a standard computation in <5 mins that would take a leading supercomputer over 10^25 years, far beyond the age of the universe(!).
This, I think, is the future: the exact UI you need — and nothing more — exactly when you need it, and at no time else.
stratechery.com/2024/the-gen…
via @stratechery & @benthompson
"Working with AI is a dialogue, not an order."
"Treat AI just like an infinitely patient new coworker who forgets everything you tell them each new conversation."
Great guidance on how to think about working-with & prompting AI systems from @emollick
oneusefulthing.org/p/getting…
Top Learning Hack:
Watch YouTube tutorials at the *fastest speed* you can handle while still following the content.
Why?
1. You get through videos faster (obviously...)
But the real magic?
2. It forces your brain to stay 100% focused on the content. No chance to daydream, just pure engagement.
Try it.
Game changer.
Enjoyed this one @martinrue
Reminded me of this graphic from @waitbutwhy
Doing More Things = More Path Options From Today
