Be Late to the Party 🎉

Get insights into real-world AI without the hype.

(Don't worry, unsubscribe anytime and I won't give your email to anyone else. I'm not Facebook.)

Understanding Machine Learning Validation eBook

⭐⭐⭐⭐⭐ (4 reviews)

"Understanding Machine Learning Validation"

Sign up and receive a complimentary copy of my e-book.

Learn how to avoid overfitting and improve model generalization, even in complicated data environments. This guide will touch on essential topics, including train-test splits, cross-validation, stratification, and spatial & temporal validation.

Build the best machine learning model you can.

Trusted by 1,111+ readers

And eyed suspiciously by Kenneth, but we'll convince him one day!

"Who are you and why do you have a microphone?!"

"I'm eyeing this newsletter suspiciously! You'll never convince me."

"Lorem Ipsum!"

Always Free

A weekly newsletter, landing in your inbox every weekend for 2 years and counting.


Get AI, science, and tech news without FOMO or unnecessary hype.

100% Unsponsored

Over 2 years of unsponsored curated tools, recommendation and news!

Typical Issue

A typical issue of Late to the Party 🎉 includes the following sections:

  • The Latest Fashion - 3 Links to great machine learning projects.
  • My Current Obsession - Relevant developments and neat things I did.
  • Thing I Like - Something that made my neurodivergent life easier.
  • Hot Off the Press - New pieces of content from around the web I created.
  • Python Deadlines - New and upcoming deadlines for
  • Machine Learning Insight - Deep Dive into the ML topic from last week.
  • Question of the Week - Challenge to think deep about an AI topic.
  • Tidbits from the Web - 3 links to content around the web I enjoyed.

Just mix and match what you want to read!

Check out a selection past issues below 👇

Smartphone example

Explore Past Issues

Issue Header Image
🦉 Birds of night are a hoot. They give it owl they've got!

How are you doing today? We’ve got some fresh, hot machine learning right from the source today!

  • CNet ran a bunch of AI-generated articles and found errors in more than 50% in hindsight
  • Pyvista looks really nice for 3D visualization in Python
  • You can write GPT from scratch in 60 lines

Visit Issue #116
Issue Header Image
🐰 I’m so egg-cited, and I just can’t hide it

I am deeply enjoying the long Easter weekend here, but I won’t deprive you of your weekly machine learning dose!

  • I loved this annotated version of the transformer paper for deeper understanding!
  • F-strings are great, but I can never remember the number formatting. Here’s a cheat sheet!
  • The Learning interpretability tool is a nifty often source platform for ML explainability

Visit Issue #121
Issue #151 Header Image
🎉 I like this newsletter in party-cular

it’s been 2 years of this newsletter! 🥳

In these anniversary issues, I like to go through the last year of writing this weekly newsletter and see what everyone liked. So, in case you missed these awesome links, this is what everyone else enjoyed!

  • The Explainer Dashboard for interactive exploration of machine learning models
  • Messing up code is easy, and Dangit, git?! to find the right command to untangle the mess
  • Everyone loved the idea of easier matplotlib subplots with mosaics.

Visit Issue #93
Issue #104 Header Image
🎨 We put the Art in Artificial Intelligence!

what an eventful week. Twitter seems to be dying slowly, and winter has eventually arrived. Let’s warm our hearts with some machine learning!

  • Facebook created a “scientific AI” called Galactica, and it was terrible and pulled after only 3 days.
  • Quantus is an explainable AI kit that uses different XAI techniques.
  • Notion has developed a writing AI, and you can sign up for the Beta Waitlist now!

Visit Issue #104
Issue #151 Header Image
🎄 Lucky I got an advent calendar, their days are numbered

The Christmas season started! Let’s look at some merry machine learning!

  • Love this idea! They published a high-quality audiobook version of the Stochastic Parrot Paper!
  • The Incredible Pytorch is a curated list of various content around Pytorch
  • Friends don’t let friends make pie charts.

Visit Issue #151

Frequently Asked Questions

Late to the party is a newsletter focused on exploring various aspects of real-world machine learning.

It includes topics like new machine learning models, courses like the scikit-learn MOOC, and Deepminds latest and greatest scientific revolution, ensuring you're equipped to build robust and effective machine learning career.

Get insights on real-world machine learning, data science, and Python every Friday from Scientist for Machine Learning at ECMWF Dr. Jesper Dramsch. 🎉

I joined Facebook, when it became trendy to delete it. I learned Python, when Javascript ate the world. I learned Tensorflow when everyone already loved Pytorch. If like me, you're usually Late To The Party, this newsletter is for you.

In this weekly newsletter, I share interesting tidbits real-world machine learning that I found around the web. Because, while often being late to the party, I love learning new stuff and I love sharing what I learn. So maybe this is something for you too.

I don't take sponsorships to sell you "links", you can read about this in my Ethics Statement

Join 1,111+ readers and get the book for free!

Each newsletter issue not only discusses theoretical aspects but also provides practical tips and examples of how these techniques can be applied in real-world scenarios, especially Earth science. This ensures that you can directly implement what you learn in your projects or work.

Yes, most issues are available in the archive!

Each issue has a little survey at the end, where you can suggest topics for future issues. You can also reply to any email directly.

I do however not use sponsorships, cover commercial releases or guest posts, so please don't ask for that.

The newsletter is free.

But there is a voluntary Premium membership. I am currently paying for the newsletter out of my own pocket, but if you want to support me, you can become a premium member. Premium members get access to the full archive.

There are no ads, no sponsorships, and no guest posts. I am not selling your data, and I am not giving it to anyone else. I am not Facebook.

You can read more about this in my ethics statement.


I want to make this information available to as many people as possible.

If you haven't received it yet, it is still on its way. It will be sent out within the first week of your free membership!

Meet the Creator

Jesper Dramsch (they/them) works at the intersection of machine learning and physical, real-world data with 8 years of experience.

Currently, they're working as a scientist for machine learning in numerical weather prediction at the coordinated organisation ECMWF.

Jesper is a fellow of the Software Sustainability Institute, creating awareness and educational resources around the reproducibility of machine learning results in applied science. Before, they have worked on applied exploratory machine learning problems, e.g. satellites and Lidar imaging on trains, and defended a PhD in machine learning for geoscience. During the PhD, Jesper wrote multiple publications and often presented at workshops and conferences, eventually holding keynote presentations on the future of machine learning in geoscience.

Moreover, they worked as consultant machine learning and Python educator in international companies and the UK government. They create educational notebooks on Kaggle applying ML to different domains, reaching rank 81 worldwide out of over 100,000 participants and their video courses on Skillshare have been watched over 128 days by over 4500 students. Recently, Jesper was invited into the Youtube Partner programme creating videos around programming, machine learning, and tech.

Jesper Dramsch is the creator of,, and the Latent Space Community.

Jesper Dramsch