Henry Garner
Principal Engineer & AI Lead @juxt.pro
AI/ML • data intensive applications • product
Author of Clojure for Data Science cljds.com/book
Maintainer of Clojure(Script) library kixi.stats
{London | Cambridge | Brighton}, UK
- Don’t let the title fool you, this is one of the most personal talks I’ve given. It goes beyond the technology: I think AI skeptics are missing genuine opportunities, while AI enthusiasts risk sacrificing the context required for mastery, satisfaction, and responsibility. Please watch and share.
- [Updated link] www.juxt.pro/blog/xt25-fi...
- New JUXTCast episode! I had a blast chatting with the team about our upcoming AI Radar. We explored practical insights beyond the hype, from inconsistent AI engineering studies to why classical ML still matters. Thoughtful perspectives to help navigate the AI ecosystem: www.juxt.pro/blog/juxtcas...
- I did a podcast! Many thanks to @juxt.pro for inviting me to talk about a project I’ve been working on for the past few months: the JUXT AI Radar 📡 Links for all the usual podcast platforms here: juxt.pro/blog/juxtcas...
- Reposted by Henry Garner[Not loaded yet]
- There are two grand goals of #AI research, and we're fixated on the wrong one.
- Ever since the earliest days of cybernetics we've been captivated by what Ben Shneiderman calls the 'Science Goal' of building systems that can replace human judgement entirely. And this long-imagined future where autonomous agents and robot teammates take over our roles may finally be upon us.
- Yet there's always been a parallel 'Innovation Goal' focused on extending human capabilities instead of replacing them. William Ross Ashby wrote about Intelligence Amplification, (IA, not AI), back in 1956.
-
View full threadSo for me, the question isn't whether AI can think like us. It's whether we can design it to help us think better.
- Another day shovelling thoughtcoal into the Claude-furnace
- The Work of Software Engineering in the Age of Mechanical Reproduceability
- There was a time when anyone could be an artist 👨🎨 — if they loved the work enough. Then came the professionals. Then came the camera. Today, it’s not paint that’s being replaced. It’s code. 9elements.com/blog/when-sk...
- Am I “ready to revamp my AI strategy for 2025?” I don’t know grandma I just wanted to drop in and give you these flowers
- Researchers recently studied two fundamentally different approaches to AI-assisted investment decision making: www.microsoft.com/en-us/resear...
- In the first approach, AI generated recommendations for investors to review. In the second, investors developed their own investment thesis first then received feedback on their reasoning from the AI.
- The difference in outcomes was striking. The second approach, where AI built on investors’ own ideas, led to better portfolio diversification, fewer but more strategic trades, and significantly higher satisfaction: 67% versus 43%.
-
View full threadAre you sure you know which are the ‘predictable and routine’ tasks which can be safely delegated? Are you practicing reliable, safe, and trustworthy #AI use?
- Lately I’ve been thinking about the Collingridge dilemma.
- When a technology first emerges we can easily shape its development. Yet we can't fully grasp its impact until it’s widely used.
- When change is easy at the beginning, the need for it can’t be foreseen. Once the need for change is apparent, it’s become difficult, expensive, and time-consuming to implement.
-
View full threadUnderstanding when and how to intervene—maintaining agency whilst leveraging AI's capabilities—is, I think, one of the defining technical skills of our age.
- Two and a half millennia ago, Aristotle popularised a word that's more relevant today than ever: phronesis (fro-NEE-sis)
- He used it to describe “practical wisdom”: the ability to navigate life's complex challenges with sound judgment. It’s the capacity to find the right balance between competing values instead of rigidly following proscribed rules.
- Over the past 50 years, this ancient term has experienced a revival. Medical breakthroughs have forced us to grapple with profound questions. When does life begin and end? How do we balance quality versus quantity? Who gets access to limited resources?
-
View full threadAs technologists using AI developer tooling, how are we maintaining our own agency and exercising our phronesis?
- The concept of thinking will surely evolve. Every time we invent a machine to do something we thought only humans could do, we discover a little bit more about what it really means to be human.
- In his book Human-Centered AI Ben Shneiderman argues that "interface designs that are consistent, predictable and controllable are comprehensible, thus enabling mastery, satisfaction and responsibility."
- Yet much of today's AI pushes us toward unpredictable prompt-based interfaces that aren't intuitive for regular users or stable enough for experts. cacm.acm.org/opinion/prom...
- Many vendors want us to think of AI as a "teammate." But this metaphor might actually limit how we design the next generation of tools. Steve Jobs once said computers could be "a bicycle for our mind”… but bicycles are nothing like human bodies. www.youtube.com/watch?v=NjIh...
- The question isn't whether AI can think for us, but whether it helps us think better. Can it scaffold our understanding whilst keeping us in control? Can it challenge our assumptions without making decisions for us? What matters isn't what AI can do on its own, it’s what it helps us achieve.
- Reposted by Henry GarnerRepeatedly humiliating my LLM with “don’t make me ask Claude” so it returns valid JSON
- No one who’s applied optical character recognition at scale will be surprised to learn that Pokemon was quite popular in the early 18th century.
- Reposted by Henry Garner[Not loaded yet]
- Reposted by Henry Garner[Not loaded yet]
- Reposted by Henry Garner[Not loaded yet]
- Reposted by Henry Garner
- Claude’s got Wife Guy energy and I’m totally here for it
- My tax return is due so the oven has never looked cleaner
