Max kuhn
banner
topepo.bsky.social
Max kuhn
@topepo.bsky.social
Writing modeling packages at @posit.co (née RStudio). Opinions are my own. https://max-kuhn.org/
Pinned
I last posted here about 6 months ago. Here's what I've been working on and/or thinking about. #rstats, #statistics, #ml

package upkeep: we are doing major preventive maintenance on the tidymodels packages ("upkeep week!"). It's rote but very rewarding work. Error messages are 100x better.

1/3
Applied Machine Learning for Tabular Data
aml4td.org
Reposted by Max kuhn
For more than a year I have been working on a brand new Jupyter Notebook editor for Positron. This is a ground-up build of a new Jupyter Notebook experience built to leverage all the knowledge and tools Posit/Positron brings to the data science table. 🧵#jupyter
February 4, 2026 at 12:54 PM
Reposted by Max kuhn
dplyr 1.2.0 is out now and we are SO excited!

- `filter_out()` for dropping rows

- `recode_values()`, `replace_values()`, and `replace_when()` that join `case_when()` as a complete family of recoding/replacing tools

These are huge quality of life wins for #rstats!

tidyverse.org/blog/2026/02...
dplyr 1.2.0
dplyr 1.2.0 fills in some important gaps in dplyr's API: we've added a new complement to `filter()` focused on dropping rows, and we've expanded the `case_when()` family with three new recoding and re...
tidyverse.org
February 4, 2026 at 11:39 AM
Reposted by Max kuhn
caretForecast
Conformal Time Series Forecasting Using Machine Learning
taf-society.github.io
February 1, 2026 at 4:41 PM
Reposted by Max kuhn
Tomorrow at the Data Science Lab 🧪 we are hearing from the amazing @theotheredgar.bsky.social about the {mall} package:

Run Natural Language Processing against your #RStats tibbles or #Python Polars DataFrames for sentiment analysis, text summaries, and more!

Join us at 12 pm ET: pos.it/dslab
January 26, 2026 at 5:26 PM
Reposted by Max kuhn
I sent 200 pull requests using Claude Code and wrote about the experience. It's pretty wild!

For dplyr releases, we send a PR any time we break an #rstats package. This release advances a lot of deprecated functions, triggering issues in many old packages!

blog.davisvaughan.com/posts/2026-0...
Semi-automating 200 Pull Requests with Claude Code – Davis Vaughan
blog.davisvaughan.com
January 12, 2026 at 1:42 PM
Reposted by Max kuhn
We are excited to see that xgboost recently had a big CRAN release! We have worked hard on the tidymodels team to make sure you all have a smooth transition.

Please yet us know if you are experiencing any issues with the releases

tidyverse.org/blog/2025/12...
#rstats #tidymodels
tidymodels & xgboost
The tidymodels ecosystem is prepared for big xgboost CRAN release.
tidyverse.org
December 15, 2025 at 5:39 PM
Reposted by Max kuhn
~~ making sense of academic statistics ~~

i wrote about the confusing relationship between statistics and data analysis, and also about how statistics relates to science

#statistics #rstats #datascience

www.alexpghayes.com/post/making-...
July 15, 2025 at 8:15 PM
Reposted by Max kuhn
I'm excited to announce the newest release of {tidypredict}!
This release brings for standardization to outputs, faster trees for parsing and prediction, and glmnet support

tidyverse.org/blog/2025/12...
#rstats #tidymodels
tidypredict 1.0.0
tidypredict 1.0.0 brings faster computations for tree-based models, more efficient tree representations, glmnet model support, and a change in how random forests are handled.
tidyverse.org
December 10, 2025 at 11:03 PM
We’ve released two new tidymodels #rstats packages for feature selection: filter and important.

tidyverse.org/blog/2025/11...
Two New tidymodels Packages
Two new tidymodels packages focus on supervised feature selection.
tidyverse.org
November 24, 2025 at 7:12 PM
Reposted by Max kuhn
I’m not aware of an Arrow or Parquet format, but there is the ONNX format (see onnx.ai). Depending on the model, you could try Posit’s orbital project, which translates your model to SQL (see here orbital.tidymodels.org).
ONNX | Home
onnx.ai
November 24, 2025 at 10:54 AM
Reposted by Max kuhn
We are excited to share a few updates happening in tidymodels in this blog post!

tidyverse.org/blog/2025/11...
Q3 2025 tidymodels digest
A summary of what has been going on for the tidymodels group in the mid 2025.
tidyverse.org
November 19, 2025 at 12:19 AM
Reposted by Max kuhn
🤖 Introducción a #Tidymodels
Aprendé a usar tidymodels para crear y evaluar modelos de #MachineLearning en #RStats
Con Francisco Cardozo y Edgar Ruiz

📅 1 Dic, 18:00–20:00 UTC-3 – Online
💵 Estudiantes USD 5 · Académicos USD 10 · Industria USD 15
🔗 www.eventbrite.com.ar/e/1962543491...

#LatinR2025
November 13, 2025 at 3:33 PM
We're hiring an open-source #python developer focused on modeling APIs!

tidyverse.org/blog/2025/11...

#numpy #scipy #scikitlearn
Python Open-Source Developer
Posit is hiring a Python open-source developer to create more data analysis tools.
tidyverse.org
November 12, 2025 at 5:46 PM
My #rstats talk for the R+AI conference, titled "Analyzing #LLM Evaluations," starts at 1:40 EST today.

rconsortium.github.io/RplusAI_webs...
November 12, 2025 at 5:38 PM
I just gave a 20m overview of the #TabPFN deep learning model at R/Pharma.

The slides are at: topepo.github.io/2025-r-pharma/

(link to Quarto files on the second slide)
November 5, 2025 at 7:10 PM
We've made a blog post about a new major version of the #rstats tune package! Two main changes: parallel processing frameworks and the ability to tune postprocessors.

tidyverse.org/blog/2025/11...
tune version 2.0.0
A new version of tune brings significant updates for model tuning and parallel processing.
tidyverse.org
November 5, 2025 at 7:06 PM
Reposted by Max kuhn
I'm excited to share side::kick(), an experimental open-source coding agent for RStudio built entirely in R. It can interact with your files, communicate with your active #rstats session, and run code.

Check it out: github.com/simonpcouch/...
November 5, 2025 at 3:57 PM
Reposted by Max kuhn
I'm excited to share a new version of the Air formatter for #rstats, with support for tabular formatting! Super useful with `tibble::tribble()` calls or `data.table::fcase()`.

It's currently experimental as we're looking to get feedback on the feature, so please let us know what you think.
October 21, 2025 at 10:34 AM
Before #NoKings in Hartford CT...
October 18, 2025 at 7:33 PM
Reposted by Max kuhn
Shannon's slides are always so unbelievably clear and helpful!!!

github.com/shannonpileg...

I'm having "Ohhhhh that's what that means" moments every 10 seconds here.
#positconf2025
September 18, 2025 at 3:09 PM
Reposted by Max kuhn
I'm all in on @topepo.bsky.social and co's new {important} and other variable importance/feature selection tools in tidymodels! #rstats
September 12, 2025 at 6:17 PM