This week's productive procrastination involved getting nerd-sniped by seeing if I could get real-time speech-to-text working in Shiny with the Speechmatics realtime API (my speechmatics #rstats 📦 published today on CRAN currently supports their batch API). This was fun though needs lots of work!

Next to Loch Ken for the first time in 20 years. Different textures of smoke. First camp of the year. Perfection.

It's a CORS thing! 🤖 suggested to use "jsDelivr - proxies GitHub repos with proper CORS headers" - and now I read the data in!

Hmm...

and `dplyr::recode_values()`, `dplyr::replace_values()`, and `dplyr::replace_when()`:

as well as `dplyr::when_any()` and `dplyr::when_all()`:

arrow now supports some of your favorite new dplyr verbs, such as filter_out():

Almost exactly 8 years after first trying it in Thailand and falling in love with the dish, finally got around to making mango with sticky rice! 😋 Crispy mung beans are *essential* to the texture and so I'm hesitant to order it in a restaurant in the UK/US as most I've been to have skipped it.

There's promise in using LLMs for code review, but it's tricky things to make sure it's not overwhelming. I was looking at this new experimental package by Simon Couch and I really love how it allows you to review code iteratively. #rstats #ai #llms github.com/simonpcouch/...

We've added support for stringr::str_ilike() for case-insensitive pattern matching.

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

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

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

Woke up to snow and abnormally cold weather for the UK; -6°C! It's a this kinda hat kinda day!

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

This was just a toy example, but in a real data extraction pipeline, I'd be better off using pre-defined categories. Here, if I use `type_enum()` in my structured output, the LLM must choose from a specific list. (2/3)

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

Most updates were to docs and internals, but one feature to note is that we’ve added support for stringr::str_replace_na()

Fantastic keynote by Wes McKinney @ #EARL2025 - personal highlights: 🧏 AI is less of a threat to data science as requires human judgment 🤔 AI becoming the new literacy 🗣️ Learn to use coding agents and AI, & articulate reviews 📖 Spend more time reading code, less time writing #ai #rstats #python

Great talk by @romainfrancois.bsky.social on how xeus-r makes it easier to work with #rstats in Jupyter; it's simpler than IRkernel and in combination with WebR opens up a ton of cool possibilities for in-browser Jupyter+R 🎉 #EARL2025

Anyone else been using Jupyter notebooks for year and didn't know the etymology of the name?! #EARL2025

Seriously cool stuff from @shikokuchuo.net on mirai+ellmer+shiny+httr2 with otel for telemetry at Shiny in Production! You can use it to improve the performance of your Shiny apps by getting really granular information about it what's going on! #rstats

A delightfully concise history of #rstats from @colinfay.bsky.social at Shiny in Production 🔥

Apache Arrow will be 10 soon! 🥳 At the Arrow summit, PMC chair @enpiar.bsky.social reflects on the keys to longevity: ⭐ Social components are as important as the technical ⭐ Build relationships with other projects & make newcomers feel welcome & empowered ⭐ Culture is what makes the difference

If you've ever gotten stuck with {renv} you need to check out Shannon Pileggi's talk at #PositConf25 - learning soooo much! #rstats

Learning some really cool stuff about running parallel purrr locally with mirai at #PositConf2025

We also now support Decimal 32 and 64 type objects

We’ve also added support for hms::hms() and hms::as_hms() for creating and working with time of day variables.

Arrow now supports hyperbolic trigonometry functions