Nic Crane
LightNews LightNews
niccrane.bsky.social
Nic Crane
@niccrane.bsky.social
2.5K followers 160 following 93 posts
Independent R consultant. Apache Arrow PMC Member & #rstats 📦 maintainer.

Arrow course launching early 2026: https://big-data-r.thinkific.com/

More of my stuff at https://niccrane.com/
Posts Replies Media Videos
niccrane.bsky.social
We're excited to announce the release of {arrow} 23.0.0 🏹📦

Here's a roundup of the new features and changes in a 🧵

Full details can be found at arrow.apache.org/docs/r/news/

#rstats #apachearrow
Changelog
arrow.apache.org
January 28, 2026 at 4:55 PM
niccrane.bsky.social
Tool calling lets LLMs run R functions; in this example I let an LLM ask my R session to check the latest {ellmer} updates by scraping the news page and when I ask the LLM "what's new in ellmer?", it works with what comes back.

{ellmer} website: ellmer.tidyverse.org

#rstats #llms #ai #datascience
First part of code - full code can be found at https://gist.github.com/thisisnic/eae09dbd4594e2cff75d156a8bab3f59 First part of code - full code can be found at https://gist.github.com/thisisnic/eae09dbd4594e2cff75d156a8bab3f59
January 27, 2026 at 6:04 PM
Reposted by Nic Crane
sara-altman.bsky.social
More on LLMs and plot interpretation: they do fine in normal conditions, but struggle when the plot conflicts strongly with their priors.

@simonpcouch.com and I investigated why and what might help: posit.co/blog/llm-plo...
Line chart showing percent correct on the y-axis and three conditions on the x-axis: Baseline, Intuitive, and Mocked. Three lines represent GPT-5.2, Claude Opus 4.5, and Gemini 2.5 Pro. All three models score between 93-98% on baseline, then drop on intuitive and mocked conditions. All three perform the worst on the mocked condition.
January 22, 2026 at 9:31 PM
niccrane.bsky.social
I love "structured output" as a way of extracting data from text as data frame. 🎯

Image shows using the {ellmer} package and how using type_array(type_object(...)) automatically returns a data frame in R 🔧

{ellmer} website: ellmer.tidyverse.org

#rstats #llms #ai #datascience
Code in which text from wikipedia article being passed into chat_structured method to extract dates and events
January 22, 2026 at 3:10 PM
Reposted by Nic Crane
ivelasq3.bsky.social
posit::conf(2026) call for talks is now open! If you're an #RStats or #Python user, have a great DS workflow to share, or have some lessons learned, we'd love to hear from you.

🔗 posit.co/blog/posit-c...
Posit::conf(2026) Call for Talks - Posit
posit::conf(2026) is coming September 14-16 to Houston, TX, and we're looking for talks!
posit.co
January 22, 2026 at 2:58 PM
niccrane.bsky.social
RAG doesn't guarantee reliability 🤔

Built a RAG chatbot for a course & tested the same question+model three times with {vitals} - one run got it wrong. ❌

This is why evals matter: catch inconsistencies you'd miss manually. 🔍

vitals.tidyverse.org

#rstats #llms #ai #datascience
vitals logo (a teddy bear with a stethoscope), above output from running evaluation on a dataset, with 2 correct responses from the LLM and 1 incorrect response
January 20, 2026 at 2:54 PM
niccrane.bsky.social
Excellent post about Claude Code's actual energy usage 💡

Most AI energy posts only look at single queries. Simon breaks down full coding sessions - much higher, but still about the same as running a dishwasher once a day.

simonpcouch.com/blog/2026-01-20-cc-impact/

#rstats #ai #llms
January 20, 2026 at 2:48 PM
Reposted by Nic Crane
rconsortium.bsky.social
R Consortium webinar: Scaling up data analysis in R with Arrow. Learn larger-than-memory workflows, why Parquet matters, and where DuckDB fits—w/ Dr Nic Crane (Arrow R maintainer; Apache Arrow PMC). Register: r-consortium.org/webinars/sca... #rstats #arrow @niccrane.bsky.social
main image for Scaling Up Data Analysis in R with Arrow R Consortium webinar
January 20, 2026 at 12:54 AM
Reposted by Nic Crane
posit.co
Speak at posit::conf(2026) and share your R & Python stories!

Accepted speakers get:
✨ Travel & lodging help
✨ Free conference pass
✨ Professional coaching

Apply by Feb 6 to join us Sept 14-16 in Houston, TX!

Submit here: pos.it/conf-talk-2026

#positconf2026 #rstats #pydata
January 15, 2026 at 3:40 PM
niccrane.bsky.social
Building LLM training materials for R users (inc a RainbowR conference workshop) and want to know what topics people care about. ~10 min survey, results shared openly!

forms.gle/uAtVwpRhKFVU...

Boosts appreciated 🤖 #rstats #llms #ai
Survey on Learning Goals for LLMs in Data
forms.gle
January 15, 2026 at 1:59 PM
niccrane.bsky.social
Woke up to snow and abnormally cold weather for the UK; -6°C! It's a this kinda hat kinda day!
Person wearing fluffy hat Sun rising over city and snow covered rooftops
January 5, 2026 at 9:45 AM
niccrane.bsky.social
Working on Arrow in Positron and PR comments are showing up inline! 😍 OK, I'd been a bit hesitant to move away from RStudio as I was used to it, but this is a game changer! I know this is a VS Code extension thing not a Positron thing, but VSCode was a bit meh for R, so rarely used it! #rstats
December 11, 2025 at 6:51 PM
Reposted by Nic Crane
libbyheeren.bsky.social
rainbowR is gonna have a conf! They have a call for talks open until Dec 8th. Submit for full or lightning talk in these categories:
- showcasing how rainbowRers use R
- working with LGBTQ+ data
conference.rainbowr.org/abstracts.html

You can follow them at @rainbowr.org ✨ #rstats #databs #rainbowR
Abstracts – rainbowR conference
conference.rainbowr.org
December 4, 2025 at 2:42 AM
Reposted by Nic Crane
jumpingrivers.com
We’re #hiring at Jumping Rivers!

Data Scientist:
Work on client projects, modelling, dashboards, and training. Looking for 2–3 years of experience in data science.

📍 Newcastle upon Tyne (UK-wide considered)

Apply: jumping-rivers.welcomekit.co/jobs/data-sc...

#TechJobs #DataScience
December 4, 2025 at 1:53 PM
niccrane.bsky.social
Before I go setting up a CI job to automate it, does anyone know if there's a reliable way of finding out when CRAN update their extra checks: cran.r-project.org/web/checks/c...? #rstats
CRAN Package Check Issue Kinds
cran.r-project.org
November 27, 2025 at 3:58 PM
niccrane.bsky.social
Wait, is Pluribus about AI? Actually, how can it not be...
November 22, 2025 at 12:03 AM
Reposted by Nic Crane
statsrhian.bsky.social
Call for abstracts for the first ever @rainbowr.org conference is still open! We're looking for talks that either

🦄 Showcase how our diverse community uses #RStats
🏳️‍🌈 Use LGBTQ+ data

Get in your abstracts before December 1st.

conference.rainbowr.org/abstracts.html
#queerdata #PrideinSTEMDay
Abstracts – rainbowR conference
conference.rainbowr.org
November 20, 2025 at 4:59 PM
Reposted by Nic Crane
etiennebacher.bsky.social
#rstats Please welcome Jarl, a new R linter.

Jarl is a CLI tool with extensions in VS Code, Positron, and Zed. It can check thousands of lines of R code in milliseconds.

Jarl provides several output formats, a Github Actions workflow, and more.

Blog post: www.etiennebacher.com/posts/2025-1...
November 20, 2025 at 4:50 PM
niccrane.bsky.social
My talk from posit::conf 2025 is now online! 🎉 It's about my early experiments in using LLMs to be more productive; what went well, what didn't, & what I learned, in the hope it can be useful to other people experimenting with AI workflows 🧪 🤖

www.youtube.com/watch?v=_nhD...

#rstats #llms #ai
Hacking Productivity with LLMs (Nic Crane, NC Data Labs) | posit::conf(2025)
YouTube video by Posit PBC
www.youtube.com
November 19, 2025 at 8:32 AM
niccrane.bsky.social
Last week I posted about using structured output in {ellmer} to turn text into data, but a friend pointed out that the LLM had mixed up the words "conscious" and "consciousness" in its summary.

So how can we fix that? 🧵(1/3)

#rstats #llms #ai
November 18, 2025 at 10:08 AM
niccrane.bsky.social
Fascinating blog post on LLM accuracy in data analysis, demonstrating how relying on prompt engineering doesn't guarantee accuracy - we still need human intervention.

When plotted data was not as expected by the LLM, even cutting edge models were wrong >90% of the time!

posit.co/blog/introdu...
When plotting, LLMs see what they expect to see - Posit
Data science agents need to accurately read plots even when the content contradicts their expectations. Our testing shows today's LLMs still struggle here.
posit.co
November 14, 2025 at 11:51 AM
niccrane.bsky.social
Really fantastic guide by Lada Kesseler about working with AI coding agents. 📖🤖

It's divided into:

- Obstacles: limitations you're likely to run into ⛔

- Anti-patterns: common mistakes people make 👎

- Patterns: strategies and best practices 🦸‍♂️

lexler.github.io/augmented-co...
Augmented Coding Patterns
A collection of emerging patterns, anti-patterns, and obstacles for effective AI-augmented software development
lexler.github.io
November 13, 2025 at 4:42 PM
niccrane.bsky.social
If you've ever fought with {renv}, you *have* to go watch Shannon's talk from posit::conf. I had soooo many 💡 moments the first time I saw it!
pipinghotdata.com Shannon Pileggi @pipinghotdata.com · Nov 11
The recording of my #rstats posit::conf talk ✨ practical {renv}✨ is available! 🎉

I created this talk for frustrated {renv} users 🫠 and potential {renv} users🤩

youtu.be/l01u7Ue9pIQ?...

Even if {renv} isn't on your radar, you may still find this talk interesting if you are curious about🤔...
Practical {renv} (Shannon Pileggi, The PCCTC) | posit::conf(2025)
YouTube video by Posit PBC
youtu.be
November 11, 2025 at 8:59 PM
niccrane.bsky.social
I'm *loving* structured outputs in {ellmer} for extracting data from text directly into R objects, removing the need to write pedantic prompts about output format! 😻

Here's a quick example - extracting info about Kendrick Lamar from his Wiki page.

gist.github.com/thisisnic/e4...

#rstats #llms
November 11, 2025 at 6:44 PM