Simon Thornewill von Essen
@sthornewillve.bsky.social
Data Scientist (Bayesian) 💻 | Go/Baduk Player (~1d) ⚪️⚫️ | Vegan 🌱 | Kettlebell/Resistance Band Enthusiast 🏋️♂️ | ADHD 🧠 | Obsidian user 🪨
The problem isn’t the bot 🤖, It’s the questions you’re asking it.
I wrote about how to stop turning LLMs into digital sycophants,
the key is to get better at disagreeing with yourself.
Read all about it here: medium.com/ai-in-plain-...
I wrote about how to stop turning LLMs into digital sycophants,
the key is to get better at disagreeing with yourself.
Read all about it here: medium.com/ai-in-plain-...
How to stop your LLM from glazing you
Prompting for Discomfort
medium.com
August 4, 2025 at 11:16 AM
The problem isn’t the bot 🤖, It’s the questions you’re asking it.
I wrote about how to stop turning LLMs into digital sycophants,
the key is to get better at disagreeing with yourself.
Read all about it here: medium.com/ai-in-plain-...
I wrote about how to stop turning LLMs into digital sycophants,
the key is to get better at disagreeing with yourself.
Read all about it here: medium.com/ai-in-plain-...
If you let AI do all the thinking, you’re outsourcing your mind 🧠, and that's no fun.
Learn the 4 principles I use while interacting with LLMs.
🚗 Stay in the driver’s seat
🔍 Know when to step back
🎯 Give it for small tasks
🧱 Provide scaffolding
Read about it here 📖:
medium.com/@sthornewill...
Learn the 4 principles I use while interacting with LLMs.
🚗 Stay in the driver’s seat
🔍 Know when to step back
🎯 Give it for small tasks
🧱 Provide scaffolding
Read about it here 📖:
medium.com/@sthornewill...
How to Use AI Without Letting It Think for You
Keeping Your Edge in the Age of LLMs
medium.com
July 13, 2025 at 7:04 PM
If you let AI do all the thinking, you’re outsourcing your mind 🧠, and that's no fun.
Learn the 4 principles I use while interacting with LLMs.
🚗 Stay in the driver’s seat
🔍 Know when to step back
🎯 Give it for small tasks
🧱 Provide scaffolding
Read about it here 📖:
medium.com/@sthornewill...
Learn the 4 principles I use while interacting with LLMs.
🚗 Stay in the driver’s seat
🔍 Know when to step back
🎯 Give it for small tasks
🧱 Provide scaffolding
Read about it here 📖:
medium.com/@sthornewill...
It's funny how I've had ADHD my whole life, but only in the recent years that I've come to understand it better, which I would like to share to raise awareness.
Read about it in my new article; sthornewillve.medium.com/what-ive-lea...
#ADHD #neurodiversity
Read about it in my new article; sthornewillve.medium.com/what-ive-lea...
#ADHD #neurodiversity
What I’ve learned about my Adult-Diagnosed ADHD from being on meds
A big challenge with any neurodivergence, is struggling to understand how present it is
sthornewillve.medium.com
May 2, 2025 at 10:53 PM
It's funny how I've had ADHD my whole life, but only in the recent years that I've come to understand it better, which I would like to share to raise awareness.
Read about it in my new article; sthornewillve.medium.com/what-ive-lea...
#ADHD #neurodiversity
Read about it in my new article; sthornewillve.medium.com/what-ive-lea...
#ADHD #neurodiversity
My cat keeps knocking over my mugs! 🐈☕ I expand upon a previous example to teach Bayesian linear modelling and how to avoid overfitting.
It’s a gentle intro to Bayesian linear modelling 📊🔍, read more about it here: sthornewillve.medium.com/how-i-modell...
#Bayesian #stats
It’s a gentle intro to Bayesian linear modelling 📊🔍, read more about it here: sthornewillve.medium.com/how-i-modell...
#Bayesian #stats
How I Modelled My Cat’s Chaos using Linear Models in PyMC
Broaden your Bayesian skills to Linear Modelling with this quick tutorial
sthornewillve.medium.com
April 25, 2025 at 10:51 AM
My cat keeps knocking over my mugs! 🐈☕ I expand upon a previous example to teach Bayesian linear modelling and how to avoid overfitting.
It’s a gentle intro to Bayesian linear modelling 📊🔍, read more about it here: sthornewillve.medium.com/how-i-modell...
#Bayesian #stats
It’s a gentle intro to Bayesian linear modelling 📊🔍, read more about it here: sthornewillve.medium.com/how-i-modell...
#Bayesian #stats
There's a lot of hype around Bayesian statistics, but learning how to use it is hard! 🥵 This post shows you how to fit the simplest possible model, calculating an average!
Take your first step into Bayesian modelling in my new article: sthornewillve.medium.com/babys-first-...
#Bayesian #stats
Take your first step into Bayesian modelling in my new article: sthornewillve.medium.com/babys-first-...
#Bayesian #stats
Baby’s First Bayesian Model Using PyMC in Python
Giving you the tools to write Bayesian models as quickly as possible!
sthornewillve.medium.com
April 18, 2025 at 6:58 PM
There's a lot of hype around Bayesian statistics, but learning how to use it is hard! 🥵 This post shows you how to fit the simplest possible model, calculating an average!
Take your first step into Bayesian modelling in my new article: sthornewillve.medium.com/babys-first-...
#Bayesian #stats
Take your first step into Bayesian modelling in my new article: sthornewillve.medium.com/babys-first-...
#Bayesian #stats
#Go / #Baduk is an ancient and mysterious game. After a decade of playing, it’s shaped not just how I think but who I am.
It taught me to challenge assumptions, develop aggression, and accept that mastery is impossible.
Read more about it here ⚫⚪: sthornewillve.medium.com/im-a-1-dan-g...
It taught me to challenge assumptions, develop aggression, and accept that mastery is impossible.
Read more about it here ⚫⚪: sthornewillve.medium.com/im-a-1-dan-g...
I’m a 1-dan Go player with a decade of exerpeince; here is what the game has taught me about myself
Go is an ancient strategy board game and is at once an elegant, enigmatic and harsh teacher.
sthornewillve.medium.com
April 12, 2025 at 6:45 PM
#Go / #Baduk is an ancient and mysterious game. After a decade of playing, it’s shaped not just how I think but who I am.
It taught me to challenge assumptions, develop aggression, and accept that mastery is impossible.
Read more about it here ⚫⚪: sthornewillve.medium.com/im-a-1-dan-g...
It taught me to challenge assumptions, develop aggression, and accept that mastery is impossible.
Read more about it here ⚫⚪: sthornewillve.medium.com/im-a-1-dan-g...
Businesses often make decisions while facing uncertainty and limited data. Bayesian methods offer a principled way to update beliefs and reduce risk. 🎯
Learn why more companies should embrace this approach in my new article: sthornewillve.medium.com/why-bayesian...
#Bayesian #stats #business #tech
Learn why more companies should embrace this approach in my new article: sthornewillve.medium.com/why-bayesian...
#Bayesian #stats #business #tech
Why Bayesian methods very useful for business decision making
Bayesian methods are very useful in a business context, but they are heavily underutilised
sthornewillve.medium.com
April 5, 2025 at 11:20 AM
Businesses often make decisions while facing uncertainty and limited data. Bayesian methods offer a principled way to update beliefs and reduce risk. 🎯
Learn why more companies should embrace this approach in my new article: sthornewillve.medium.com/why-bayesian...
#Bayesian #stats #business #tech
Learn why more companies should embrace this approach in my new article: sthornewillve.medium.com/why-bayesian...
#Bayesian #stats #business #tech
When I transitioned from chemistry to data science 7 years ago, I thought mastering the technical side was the key to success 🎓📊 but I was still missing on some key skills.
I break down the most important things I’ve learned in my latest Medium article:
sthornewillve.medium.com/these-are-th...
I break down the most important things I’ve learned in my latest Medium article:
sthornewillve.medium.com/these-are-th...
These are the most important data science skills I learned since boot camp
Important lessons that you won’t learn online but while on the job
sthornewillve.medium.com
March 29, 2025 at 9:18 AM
When I transitioned from chemistry to data science 7 years ago, I thought mastering the technical side was the key to success 🎓📊 but I was still missing on some key skills.
I break down the most important things I’ve learned in my latest Medium article:
sthornewillve.medium.com/these-are-th...
I break down the most important things I’ve learned in my latest Medium article:
sthornewillve.medium.com/these-are-th...
Writing and note-taking will 10x your productivity. 📝
I've seen huge benefits from using Obsidian to organize my thoughts and produce higher-quality work: sthornewillve.medium.com/writing-and-...
I've seen huge benefits from using Obsidian to organize my thoughts and produce higher-quality work: sthornewillve.medium.com/writing-and-...
Writing and note-taking will 10x your productivity
Why capturing your thoughts will make you faster, clearer, and more creative
sthornewillve.medium.com
March 19, 2025 at 9:20 AM
Writing and note-taking will 10x your productivity. 📝
I've seen huge benefits from using Obsidian to organize my thoughts and produce higher-quality work: sthornewillve.medium.com/writing-and-...
I've seen huge benefits from using Obsidian to organize my thoughts and produce higher-quality work: sthornewillve.medium.com/writing-and-...
Reposted by Simon Thornewill von Essen
Election officials are increasingly reliant on computer algorithms to help verify mail-in ballots. But does the use of algorithmic tools impact voters’ confidence in elections? My colleagues and I explored this (and more) in our new piece for PRQ. Here are some key highlights: 1/5🧵
lnkd.in/eNr34x-E
lnkd.in/eNr34x-E
January 25, 2025 at 3:32 PM
Election officials are increasingly reliant on computer algorithms to help verify mail-in ballots. But does the use of algorithmic tools impact voters’ confidence in elections? My colleagues and I explored this (and more) in our new piece for PRQ. Here are some key highlights: 1/5🧵
lnkd.in/eNr34x-E
lnkd.in/eNr34x-E
This was a big one for me, still is probably.
If your goal this year is to read more books, remember:
you don’t have to finish every book you start. If you’re halfway through and can’t stand the characters or just don’t care what happens, it’s okay to put it down.
Life’s too short to read books you don’t love!
you don’t have to finish every book you start. If you’re halfway through and can’t stand the characters or just don’t care what happens, it’s okay to put it down.
Life’s too short to read books you don’t love!
January 2, 2025 at 11:43 AM
This was a big one for me, still is probably.
Happy new year! I wish you all moisturisation and victory
December 31, 2024 at 11:25 PM
Happy new year! I wish you all moisturisation and victory
It's only recently that I understand that this is a understated benefit of Bayesian stats.
Perhaps it's a lot of effort for problems that don't matter that much, but if there is sufficient anxiety surrounding a decision then being able to explain how you get to your conclusions is vital.
Perhaps it's a lot of effort for problems that don't matter that much, but if there is sufficient anxiety surrounding a decision then being able to explain how you get to your conclusions is vital.
Bayesian inference (and mathematical reasoning more generally) isn’t just about getting the answer; it’s also about clarifying the mapping from assumptions to inference to decision.
statmodeling.stat.columbia.edu/2024/12/28/b...
statmodeling.stat.columbia.edu/2024/12/28/b...
Bayesian inference (and mathematical reasoning more generally) isn’t just about getting the answer; it’s also about clarifying the mapping from assumptions to inference to decision. | Statistical Mod...
statmodeling.stat.columbia.edu
December 28, 2024 at 4:50 PM
It's only recently that I understand that this is a understated benefit of Bayesian stats.
Perhaps it's a lot of effort for problems that don't matter that much, but if there is sufficient anxiety surrounding a decision then being able to explain how you get to your conclusions is vital.
Perhaps it's a lot of effort for problems that don't matter that much, but if there is sufficient anxiety surrounding a decision then being able to explain how you get to your conclusions is vital.
One thing that I noticed just now is that the bsky notification icon turns blue instead of the usual maximally-triggering horrible red colour.
I love that, but why wasn't the evil red one the default option?
I love that, but why wasn't the evil red one the default option?
December 20, 2024 at 2:10 PM
One thing that I noticed just now is that the bsky notification icon turns blue instead of the usual maximally-triggering horrible red colour.
I love that, but why wasn't the evil red one the default option?
I love that, but why wasn't the evil red one the default option?
So, #pkm is overloaded as a hashtag with "personal knowledge management" conent and Pokemon.
And while pokemon is great, I want to be a different kind of nerd. How can we resolve this tension? 😂
And while pokemon is great, I want to be a different kind of nerd. How can we resolve this tension? 😂
December 19, 2024 at 10:41 PM
So, #pkm is overloaded as a hashtag with "personal knowledge management" conent and Pokemon.
And while pokemon is great, I want to be a different kind of nerd. How can we resolve this tension? 😂
And while pokemon is great, I want to be a different kind of nerd. How can we resolve this tension? 😂
One thing that I underestimated is how Twitter-like social media forums promote good writing by forcing you to compress ideas. It may be obvious to some, but this is the most I've used a site like this.
I wonder if there is a way to encourage this when writing other docs to improve writing there.
I wonder if there is a way to encourage this when writing other docs to improve writing there.
December 7, 2024 at 11:02 AM
One thing that I underestimated is how Twitter-like social media forums promote good writing by forcing you to compress ideas. It may be obvious to some, but this is the most I've used a site like this.
I wonder if there is a way to encourage this when writing other docs to improve writing there.
I wonder if there is a way to encourage this when writing other docs to improve writing there.
The thing that I missed the most about Twitter was Data Science and Statistics Twitter. I can’t describe how happy I am to have access to this community again on BlueSky.
December 2, 2024 at 5:14 PM
The thing that I missed the most about Twitter was Data Science and Statistics Twitter. I can’t describe how happy I am to have access to this community again on BlueSky.
My face when I went to the local library during lunch the other day.
I walked in, looked incredibly awkward, I’m not sure if anyone noticed I was out of place, and then I left.
I walked in, looked incredibly awkward, I’m not sure if anyone noticed I was out of place, and then I left.
a group of fish in plastic bags with the words now what written on the bottom
ALT: a group of fish in plastic bags with the words now what written on the bottom
media.tenor.com
December 1, 2024 at 7:02 PM
My face when I went to the local library during lunch the other day.
I walked in, looked incredibly awkward, I’m not sure if anyone noticed I was out of place, and then I left.
I walked in, looked incredibly awkward, I’m not sure if anyone noticed I was out of place, and then I left.
I just finished reading How Minds Change a couple days ago on recommendation from @hankgreen.bsky.social and I was surprised how much it changed my perspective on how to have good discussions.
Not only will it help me to do my job better as a Data Scientist, but it also mended a rift in my soul.
Not only will it help me to do my job better as a Data Scientist, but it also mended a rift in my soul.
November 30, 2024 at 9:05 AM
I just finished reading How Minds Change a couple days ago on recommendation from @hankgreen.bsky.social and I was surprised how much it changed my perspective on how to have good discussions.
Not only will it help me to do my job better as a Data Scientist, but it also mended a rift in my soul.
Not only will it help me to do my job better as a Data Scientist, but it also mended a rift in my soul.
Wow BlueSky is really taking off!
November 30, 2024 at 9:02 AM
Wow BlueSky is really taking off!
This might seem obvious but unironically this is how a lot of data analysis is done.
I should save this example for when I’m communicating this with both technical and non-technical people.
I should save this example for when I’m communicating this with both technical and non-technical people.
Every time someone makes a causal claim based on a single plot, a kitten dies
.or "The Truth is Simple"
The complexity of the description of a dog resting in a position that isn't clearly captured by common-sense terms forces us to use more words to accurately convey the scene
1/ 👇🏼
#CausalSky
.or "The Truth is Simple"
The complexity of the description of a dog resting in a position that isn't clearly captured by common-sense terms forces us to use more words to accurately convey the scene
1/ 👇🏼
#CausalSky
November 29, 2024 at 1:18 PM
This might seem obvious but unironically this is how a lot of data analysis is done.
I should save this example for when I’m communicating this with both technical and non-technical people.
I should save this example for when I’m communicating this with both technical and non-technical people.