Lightnews — Scholar-powered news

Tomasz Kalinowski

@t-kalinowski.bsky.social

2K followers 950 following 33 posts

Software Engineer at Posit PBC. I mostly post about R, Python, and Deep Learning. Github: https://github.com/t-kalinowski

github.com

Posts Media Videos Starter Packs

Pinned

Tomasz Kalinowski @t-kalinowski.bsky.social · Jun 20

I’m very pleased to announce that a preview of the 3rd Edition of Deep Learning with R is now available! mng.bz/X7xa

The book is 50% off through July 8!

Deep Learning with R, Third Edition

Deep learning from the ground up using R and the powerful Keras library!</b> Deep Learning with R, Third Edition</i> introduces deep learning from scratch with examples that use the R language and th...

mng.bz

19 100

Tomasz Kalinowski @t-kalinowski.bsky.social · 12d

For anyone who wants to try Codex Cloud for R package development, here is a setup script you can use:

gist.github.com/t-kalinowski...

Setup codex-universal (Codex Cloud) for R package development

Setup codex-universal (Codex Cloud) for R package development - setup-codex-universal.sh

gist.github.com

Reposted by Tomasz Kalinowski

Sharon Machlis @smachlis.bsky.social · 20d

The {ragnar} #RStats 📦 has a new function to inspect a data store: ragnar_store_atlas() , which visualizes the store, @t-kalinowski.bsky.social tells #Positconf2025

1 13

Reposted by Tomasz Kalinowski

Sharon Machlis @smachlis.bsky.social · Sep 6

My Workshop for Ukraine is THIS THURSDAY, 11 Sept! Donate to help medics save lives in Ukraine *and* learn how to query text data with R and the {ragnar} #RStats 📦
Can't make it from noon to 2 ET? Register and you'll get access to recording and materials. Thank you! ❤️
www.machlis.com/workshop/

Sharon's Workshop for Ukraine

Help medics save lives in Ukraine. Learn a skill or sponsor a student & support a cause.

www.machlis.com

1 4 4

Reposted by Tomasz Kalinowski

Vincent Arel-Bundock @vincentab.bsky.social · Aug 31

I'd love feedback on this #marginaleffects WIP.

In GLMs with many parameters, the marginaleffects 📦 can get a 5–20x speedup (and better SEs) by calling JAX for automatic differentiation.

Try the instructions at this link and let me know if you run into trouble.

Thanks!

github.com/vincentarelb...

5 7 38

Reposted by Tomasz Kalinowski

Dariia Mykhailyshyna @dariia.bsky.social · Aug 29

❗️Our next workshop will be on September 11th, 6 pm CEST, on RAG in R with ragnar by @smachlis.bsky.social

Register or sponsor a student by donating to support Ukraine!
Details: bit.ly/3wBeY4S
Please share!
#AcademicSky #EconSky #RStats

1 10 13

Reposted by Tomasz Kalinowski

Andrew Holz @andrewholz.bsky.social · Aug 27

Finally got Glimpse out focused on what the Posit open source team has been up to. Turns out its quite a bit! posit.co/blog/posit-g...
I continue to be happy with the approach we (Posit) are taking with LLMs and AI. Too impactful to ignore but focus on useful and safe as much as possible.

Posit

Posit's latest glimpse newsletter highlights new product releases, including the Positron IDE and several new LLM-powered tools.

posit.co

6 14

Tomasz Kalinowski @t-kalinowski.bsky.social · Aug 26

New release of quickr v0.2.0 on CRAN #rstats!

New: runif(), more control flow, more ops, many many bug fixes.

Also, I've been experimenting with chatgpts codex (cloud) in quickr. It's pretty useful! I added some setup scripts to make it easier to get started.

changelog: github.com/t-kalinowski...

1 7

Tomasz Kalinowski @t-kalinowski.bsky.social · Aug 21

ragnar v0.2.1 is now on CRAN.

It's full of small improvements! The most impactful is that registered retrieval tools now never return previously seen chunks, giving LLMs a robust ability to do deep searches.

announcement: www.tidyverse.org/blog/2025/08...
changelog: ragnar.tidyverse.org/news/

ragnar 0.2

ragnar 0.2 introduces a tidy, transparent toolkit for building Retrieval-Augmented Generation (RAG) pipelines in R.

www.tidyverse.org

1 4

Reposted by Tomasz Kalinowski

Mitchell O'Hara-Wild @handle.invalid · Aug 4

There's so many parallel sessions at #JSM2025 that it's hard to choose where to go, so I web-scraped the schedule and made a chatbot to help.

I've made it public, so hopefully it can help you too - try it out 👉 shiny.mitchelloharawild.com/jsm2025/

A screenshot of my JSM 2025 Agenda Chat bot, featuring the answer to "when is Mitchell O'Hara-Wild presenting".

5 9 19

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 22

This would be a prime example to showcase quickr!

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 22

ragnar should support this if you provide a custom `embed` function, e.g., using a local graph embedding model via reticulate, or if you just insert chunks with precomputed embeddings. This would be a good example for us to add to docs! Feel free to open a github issue if you run into any problems!

1 2

Reposted by Tomasz Kalinowski

John Muschelli @strictlystat.bsky.social · Jul 9

The @posit.co new reticulate package with managed/uv python implementation really changed the game (and speed) of using modules without a huge condaenv overhead. In scripts but also big for packages with python deps.

1 2

Reposted by Tomasz Kalinowski

Sharon Machlis @smachlis.bsky.social · Jul 17

RAG has come to the tidyverse! Check out my step-by-step #InfoWorld tutorial on how to use the {ragnar} and {ellmer) #RStats 📦s to answer natural language questions about a document or Web page.
www.infoworld.com/article/4020...

#GenAI #LLMs

How to create your own RAG applications in R

See how to query documents using natural language, LLMs, and R—including dplyr-like filtering on metadata. Plus, learn how to use an LLM to extract structured data for text filtering.

www.infoworld.com

3 11 26

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 17

It works with shiny! We’ve deployed ragnar on a few connect apps already.
Some models are more or less eager to call tools; it depends on the model. If you’re seeing an error, please open a github issue!

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 15

Other highlights:

- `read_as_markdown()` handles HTML better, now supports selectors
- Store inspector now displays chunk context and looks nicer.
- More embedding providers (Google, Databricks, Bedrock)

Full changelog: ragnar.tidyverse.org/news/

Changelog

ragnar.tidyverse.org

1 1 5

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 15

#ragnar 0.2.0 is on CRAN #rstats!

It has a MUCH improved chunker; markdown_chunk() picks better boundaries, builds context, segments by headings, and handles overlapping chunks. Oh, and ragnar_retrieve() can deoverlap retrieved chunks now.

Website got a big update too: ragnar.tidyverse.org

Retrieval-Augmented Generation (RAG) Workflows

Provides tools for implementing Retrieval-Augmented Generation (RAG) workflows with Large Language Models (LLM). Includes functions for document processing, text chunking, embedding generation, storag...

ragnar.tidyverse.org

4 16 41

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 14

Can’t believe I forgot to share this back in February... Check out this intro to RAG by @christophscheuch.bsky.social! blog.tidy-intelligence.com/posts/rapid-...

RAG before ragnar!

It’s also a good intro to ragnar internals, since ragnar makes similar implementation choices.

2 8

Tomasz Kalinowski @t-kalinowski.bsky.social · Jul 6

You might enjoy this (even if you don’t use R): happygitwithr.com

Let’s Git started | Happy Git and GitHub for the useR

Using Git and GitHub with R, Rstudio, and R Markdown

happygitwithr.com

3 2 8

Reposted by Tomasz Kalinowski

Jonathan Carroll @jonocarroll.fosstodon.org.ap.brid.gy · Jun 29

New post: https://jcarroll.com.au/2025/06/29/counting-digits-quickly/

What if you could just wave a magic wand over your R #rstast :rstats: code and have it transform into something that ran as fast as or faster than C? @t_kalinowski's {quickr} 'R to Fortran Transpiler' does that for you! […]

Original post on fosstodon.org

fosstodon.org

8 13

Tomasz Kalinowski @t-kalinowski.bsky.social · Jun 20

I’m very pleased to announce that a preview of the 3rd Edition of Deep Learning with R is now available! mng.bz/X7xa

The book is 50% off through July 8!

Deep Learning with R, Third Edition

mng.bz

19 100

Tomasz Kalinowski @t-kalinowski.bsky.social · Jun 10

Easiest way is probably to pin all package versions to a known working date:

py_require(exclude_newer = "2025-06-09")

1 1

Reposted by Tomasz Kalinowski

hrbrmstr 🇺🇦 🇬🇱 🇨🇦 🏳️‍🌈 @hrbrmstr.dev · Jun 3

#RStats WHAT MAGICKS IS THIS?!?!?! github.com/t-kalinowski...

GitHub - t-kalinowski/quickr: R to Fortran Transpiler

R to Fortran Transpiler. Contribute to t-kalinowski/quickr development by creating an account on GitHub.

github.com

5 12 32

Tomasz Kalinowski @t-kalinowski.bsky.social · May 30

We recently added a new "Getting Started" guide ragnar.tidyverse.org/articles/rag...

Blog post coming soon!

1 2 6

Tomasz Kalinowski @t-kalinowski.bsky.social · May 28

I wonder what happens if you train with a regularizer that penalizes super weights like this. I’m guessing the model quantizes better, maybe becomes more robust, or makes more efficient use of weights.

Reposted by Tomasz Kalinowski

Posit @posit.co · May 27

Introducing the btw package for teaching LLM chat apps about your #RStats package!

Inject "invisible" messages into chats via system prompts and use tool calls to dynamically fetch context when needed.

Check out a dplyr example and learn more in @simonpcouch.com's post! posit.co/blog/custom-...

A digital illustration on a pink background with white dots. A raven with a piece of paper in its beak in a colorful hexagon covered in various graphs and doodles. Dotted lines extend from the hexagon to four white rectangular documents with horizontal lines representing text. An 'AI' icon is in the upper right corner

2 12 42