Henry Garner
@hendroid.io
Principal Engineer & AI Lead @juxt.pro
AI/ML • data intensive applications • product
Author of Clojure for Data Science https://cljds.com/book
Maintainer of Clojure(Script) library kixi.stats
{London | Cambridge | Brighton}, UK
AI/ML • data intensive applications • product
Author of Clojure for Data Science https://cljds.com/book
Maintainer of Clojure(Script) library kixi.stats
{London | Cambridge | Brighton}, UK
Ben Shneiderman, author of the book 'Human-Centered AI', argues that the choice between human control and computer automation is a false dichotomy. They are actually orthogonal dimensions creating four distinct regions.
June 25, 2025 at 7:21 AM
Ben Shneiderman, author of the book 'Human-Centered AI', argues that the choice between human control and computer automation is a false dichotomy. They are actually orthogonal dimensions creating four distinct regions.
Phronesis is making a comeback, baby!
June 20, 2025 at 8:17 AM
Phronesis is making a comeback, baby!
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."
June 18, 2025 at 6:51 AM
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."
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.
April 5, 2025 at 6:40 PM
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.
I’m just going to leave this here
January 7, 2025 at 8:42 PM
I’m just going to leave this here
Words showing the largest unexplained statistically-significant increase in the scientific literature 2024 vs 2020
December 17, 2024 at 4:41 PM
Words showing the largest unexplained statistically-significant increase in the scientific literature 2024 vs 2020
December 13, 2024 at 9:04 AM
I’ll be adding “respects the laws of physics” to all my specifications from now on
learn.microsoft.com/en-us/azure/...
learn.microsoft.com/en-us/azure/...
December 12, 2024 at 10:18 PM
I’ll be adding “respects the laws of physics” to all my specifications from now on
learn.microsoft.com/en-us/azure/...
learn.microsoft.com/en-us/azure/...
Always good to be reminded of BARTScore arxiv.org/abs/2106.11520
December 10, 2024 at 9:54 PM
Always good to be reminded of BARTScore arxiv.org/abs/2106.11520
Beyond cookie-cutter evals: Eugene’s rigorous and accessible into to custom AI assessment metrics is an excellent read eugeneyan.com/writing/evals/
December 10, 2024 at 9:32 PM
Beyond cookie-cutter evals: Eugene’s rigorous and accessible into to custom AI assessment metrics is an excellent read eugeneyan.com/writing/evals/
Shiny! The newly released Llama 3.3 LLM leads the LiveBench ranking for instruction following¹, beating Claude 3.5, GPT-4o, OpenAI o1, and you can run it on your local² machine.
> ollama run llama3.3
livebench.ai#/?IF=as
> ollama run llama3.3
livebench.ai#/?IF=as
December 9, 2024 at 8:50 PM
Shiny! The newly released Llama 3.3 LLM leads the LiveBench ranking for instruction following¹, beating Claude 3.5, GPT-4o, OpenAI o1, and you can run it on your local² machine.
> ollama run llama3.3
livebench.ai#/?IF=as
> ollama run llama3.3
livebench.ai#/?IF=as
Me reading the comments in your code
December 6, 2024 at 12:11 PM
Me reading the comments in your code
Mind blown by this impressive paper using genetic algorithms to discover more energy-efficient LLM architectures arxiv.org/abs/2411.17800
December 4, 2024 at 8:26 PM
Mind blown by this impressive paper using genetic algorithms to discover more energy-efficient LLM architectures arxiv.org/abs/2411.17800
Catbench: definitely the cutest way to develop your embedding vector intuition github.com/tanelpoder/c... #AI
November 29, 2024 at 8:33 PM
Catbench: definitely the cutest way to develop your embedding vector intuition github.com/tanelpoder/c... #AI
This, but for AI prompt-to-SQL tools and little Bobby Tables
November 29, 2024 at 4:17 PM
This, but for AI prompt-to-SQL tools and little Bobby Tables
You automate evaluation of your LLM outputs, right?
November 27, 2024 at 7:39 PM
You automate evaluation of your LLM outputs, right?
The ‘strangler pattern’ is a popular way to modernise legacy systems incrementally.
Success relies on corresponding organisational and cultural shifts, but I don’t know of a similarly apt name to describe these. Do you?
‘Agile’ seems too broad to be useful.
martinfowler.com/bliki/Strang...
Success relies on corresponding organisational and cultural shifts, but I don’t know of a similarly apt name to describe these. Do you?
‘Agile’ seems too broad to be useful.
martinfowler.com/bliki/Strang...
November 26, 2024 at 8:10 AM
The ‘strangler pattern’ is a popular way to modernise legacy systems incrementally.
Success relies on corresponding organisational and cultural shifts, but I don’t know of a similarly apt name to describe these. Do you?
‘Agile’ seems too broad to be useful.
martinfowler.com/bliki/Strang...
Success relies on corresponding organisational and cultural shifts, but I don’t know of a similarly apt name to describe these. Do you?
‘Agile’ seems too broad to be useful.
martinfowler.com/bliki/Strang...
As a systems engineer, I seek out simple designs as a prerequisite for reliability.
Users do care about reliability, but sometimes they care about being able to run SimCity more.
www.joelonsoftware.com/2000/05/24/s...
Users do care about reliability, but sometimes they care about being able to run SimCity more.
www.joelonsoftware.com/2000/05/24/s...
November 25, 2024 at 10:28 PM
As a systems engineer, I seek out simple designs as a prerequisite for reliability.
Users do care about reliability, but sometimes they care about being able to run SimCity more.
www.joelonsoftware.com/2000/05/24/s...
Users do care about reliability, but sometimes they care about being able to run SimCity more.
www.joelonsoftware.com/2000/05/24/s...
Highly recommend you read this beautiful, thought-provoking & funny article on contraptions (and other helicopter-like things) contraptions.venkateshrao.com/p/contraptio...
November 25, 2024 at 1:18 PM
Highly recommend you read this beautiful, thought-provoking & funny article on contraptions (and other helicopter-like things) contraptions.venkateshrao.com/p/contraptio...
Research into advanced AI reasoning mostly just proves why we all have anxiety
arxiv.org/pdf/2411.14405
arxiv.org/pdf/2411.14405
November 23, 2024 at 10:42 PM
Research into advanced AI reasoning mostly just proves why we all have anxiety
arxiv.org/pdf/2411.14405
arxiv.org/pdf/2411.14405
Thinking about this taxonomy of AI failures from the “Fallacy of AI Functionality” paper.
The first question I ask when thinking about a new system: “who’s going to get hurt when this breaks, and how badly?”
arxiv.org/abs/2206.09511
The first question I ask when thinking about a new system: “who’s going to get hurt when this breaks, and how badly?”
arxiv.org/abs/2206.09511
November 22, 2024 at 4:36 PM
Thinking about this taxonomy of AI failures from the “Fallacy of AI Functionality” paper.
The first question I ask when thinking about a new system: “who’s going to get hurt when this breaks, and how badly?”
arxiv.org/abs/2206.09511
The first question I ask when thinking about a new system: “who’s going to get hurt when this breaks, and how badly?”
arxiv.org/abs/2206.09511
Thanks for reminding me of this gem
November 21, 2024 at 11:36 AM
Thanks for reminding me of this gem
It doesn’t matter how many years of experience I gain
November 20, 2024 at 4:09 PM
It doesn’t matter how many years of experience I gain
Claude’s sensory overload is so relatable
November 19, 2024 at 10:08 PM
Claude’s sensory overload is so relatable
