Workshops at posit::conf(2026)
We’re thrilled to announce that workshop registration for posit::conf(2026) is now open! This year, we have an exciting lineup of in-person workshops as well as a virtual option. All workshops will be held on September 14, 2026. In-person workshops are full-day workshops (9 am – 5 pm), and the virtual workshop will be repeated three times throughout the date to accommodate as many time zones as possible.
Whether you’re looking to level up your Quarto skills, dive into machine learning, explore programming with LLMs, or modernize your R workflow, we have something for everyone. Read on to learn about this year’s workshop offerings.
In-person workshops
One-day workshops held in Houston, TX, on September 14, 2026.
Level up your Quarto: Patterns, templates, and tooling
Instructors:
This workshop is designed for Quarto users who want to update and extend their Quarto mastery. Led by Charlotte Wickham and Mine Çetinkaya-Rundel, this workshop will introduce new features of Quarto as well as blend practical recipes with conceptual understanding, so you can both copy-paste what works and understand why it works. Rather than a linear tour of Quarto features, the workshop is organized as a collection of focused, reusable patterns—each one solving a real authoring or publishing problem you’re likely to encounter.
Themes will include:
- Design once, use anywhere: Learn how to work effectively with light and dark modes, brands, and renderings, and how to share a consistent brand across multiple Quarto projects without duplication or fragility.
- Build reusable structures: Use template partials to factor out repeated content and layouts, making large or long-lived projects easier to maintain and evolve.
- Understand why Typst matters: Explore why Typst deserves a place on your radar, with concrete “quick wins” that show how it can dramatically simplify PDF generation.
- Work productively in Positron: Pick up Positron- and Quarto-specific productivity tips, including auto-formatting with Air, and practical ways to incorporate LLMs into your Quarto workflow.
- Master freezing and reproducibility: Demystify freeze—when to check it into version control, when to refresh it, and how to use it intentionally to balance reproducibility, performance, and collaboration.
By the end of the workshop, you’ll have a curated set of Quarto recipes and a mental model for how the pieces fit together. You’ll leave equipped to create more polished documents, scale your workflows across projects and teams, and make deliberate, informed choices about tools, formats, and authoring strategies.
Programming with LLMs in R and Python
Instructors:
Large Language Models (LLMs) offer developers unprecedented programmatic capabilities. Led by Garrick Aden-Buie and Sara Altman, this workshop will introduce ellmer (R) and chatlas (Python), Posit packages that simplify LLM API integration, handling conversation complexities, and enabling seamless interactions with AI models. You’ll gain a conceptual understanding of both ecosystems, but you can choose your own adventure (R or Python) for hands-on coding sessions.
Participants will explore system prompt design, token management, and tool calling while building familiarity with current AI technologies. The workshop demonstrates that:
- Coding with LLMs unlocks possibilities beyond standard tools
- No advanced AI background is required
- Implementing AI can be both accessible and exciting
Designed for AI newcomers and experienced developers alike, the session covers LLM integration, Shiny web app development, and touches on advanced topics like dynamic information retrieval, context engineering, and agentic workflows. The workshop will also feature a Q&A discussion with Joe Cheng on AI tooling for data science. Attendees will gain hands-on experience through guided exercises, providing the confidence to start their LLM journey.
Practical machine learning with tidymodels
Instructors:
Learn how to successfully develop Machine Learning using R. tidymodels is a complete framework that implements modeling best practices using tidyverse principles. The tidymodels framework is a collection of packages for modeling and machine learning using tidyverse principles.
Led by Max Kuhn and Emil Hvitfeldt, this workshop will teach you core tidymodels packages to preprocess data, specify and train models, and optimize tuning parameters. You’ll learn tidymodels syntax as well as the process of predictive modeling for tabular data. Time permitting, you’ll be introduced to basic pre-processing with recipes.
This workshop is for you if you:
- Are comfortable using tidyverse packages to read data into R, transform and reshape data, and make a variety of graphs, and
- Have some experience with modeling, such as linear regression, logistic regression, or tree-based models.
While the workshop will teach some terminology and modeling philosophy, it is primarily focused on showing users how to effectively develop, iterate, and validate a machine learning model.
Our modern R workflow (ft. Positron and AI)
Instructors:
This workshop is designed for data scientists who are comfortable with using R for data analysis and ready to level up their software engineering skills. Led by Hadley Wickham and Jenny Bryan, you’ll learn how to write more robust, maintainable R code while taking advantage of modern tooling (cough AI cough) that can dramatically accelerate your productivity.
We’ll cover three interconnected themes. First, you’ll learn contemporary R development practices—writing functions, organizing code into packages, testing, and documentation strategies that scale from personal projects to team collaboration. Second, we’ll explore how AI assistants like Claude Code can help you write better R code faster. Third, we’ll share practical tips for getting the most out of Positron, including features specifically designed to make R development and data exploration more fluid.
Prerequisites: Basic familiarity with R programming (e.g., using the tidyverse). You should have written a few functions before, but you don’t need any experience with package development, AI coding tools, or Positron. If you have taken a package development workshop before, you’ll learn what’s changed in our workflows, especially how we’re using AI and Positron effectively.
Modern data science in Python
Instructors:
As the data science landscape evolves, many practitioners are looking to expand their toolkit to include Python. Led by Jeroen Janssens and Michael Chow, this workshop will provide a practical, hands-on introduction to an end-to-end Python workflow.
We will move beyond the basics of syntax to focus on a modern stack that prioritizes readability and performance. Using Positron as our development environment, we’ll navigate a complete project lifecycle.
The workshop covers:
- Data Manipulation: Performant data wrangling using Polars.
- Visualization: Implementing the Grammar of Graphics in Python with Plotnine.
- Reporting: Creating presentation-ready tables with Great Tables.
- Modeling: Building machine learning models using Scikit-Learn.
The workshop is structured around a single case study. You’ll work through a series of exercises designed to help you get familiar with these tools. By the end of the day, you’ll have a good idea of how to translate your data science skills into a Python context and a repository of code examples to apply to future projects.
This workshop is a great fit for those getting started with data science as well as R users who want to get their feet wet with Python for doing data science.
A Practitioners Guide to Scalable Open-Source in Pharma with Posit Products
Instructors:
- Josef Hartmann, Boehringer Ingelheim
- Michael Mayer
- Matthias Trampisch, Boehringer Ingelheim
- Kevin Kunzmann, Boehringer Ingelheim
Building enterprise‑grade data science platforms is challenging – especially in regulated industries such as pharmaceuticals. This workshop is designed for practitioners who create or operate large‑scale deployments of Posit professional products and who need to integrate open‑source statistical computing into validated environments.
The team will present a cloud‑native architecture that brings R into a multilingual statistical computing ecosystem, building on Posit Workbench, Posit Connect, and Posit Package Manager. The architecture scales to hundreds of data scientists and supports multiple execution environments. It features high‑performance compute capabilities in Posit Workbench through a fully integrated Slurm scheduler, batch processing, scalable interactive Shiny applications on Posit Connect, and AI‑assisted workflows through Positron Assistant.
A central element of the workshop is our approach to building reproducible, containerized execution environments that behave consistently across Workbench and Connect. We discuss quality‑assurance considerations, including validation strategies suitable for regulated settings. Finally, we demonstrate how Posit products integrate seamlessly with surrounding enterprise systems and provide practical guidance on development, deployment, and lifecycle management in large organizations.
Modern Clinical Reporting in R with the Pharmaverse
Instructors:
Modern clinical reporting workflows are evolving fast, and this workshop helps you keep pace with a practical, end-to-end approach in R. You’ll learn how to move from raw data to analysis deliverables using Pharmaverse tools, including standards-aligned SDTM preparation, efficient creation of ADaM datasets, and generation of high-quality tables, listings, and graphs (TLGs) supported by Analysis Results Datasets (ARDs). You’ll also see where AI-assisted tooling can fit naturally into the workflow to support faster iteration and reduce manual rework, without sacrificing transparency, reproducibility, or reviewability. Leave with a modular, reusable playbook for applying R to every stage of clinical reporting across your team.
Virtual workshops
Half-day workshops at the virtual conference on September 14, 2026, held at three different time periods:
- Session 1 – Europe: 3:00 – 6:00 am CST, taught by François Michonneau
- Session 2 – US East: 8:00 – 11:00 am CST, taught by Garrett Grolemund
- Session 3 – US West & ANZ: 4:00 – 7:00 pm CST, taught by Garrett Grolemund
Getting started with Positron: An AI-ready polyglot workflow for data science
This virtual three-hour workshop provides a quick start for R and Python users who are curious about Positron, Posit’s new AI-ready polyglot code editor for data science. Here, you will learn how to use Positron’s time-saving features to complete a typical end-to-end data science task.
Along the way, you will learn how to:
- Install packages and set up a reproducible environment
- Connect to a database
- Add Positron Extensions
- Navigate the Command Palette
- Configure Positron to use your favorite LLM model provider
- Write, edit, and debug code with Positron’s built-in AI client
- Use an LLM to explain and translate code
- Learn the best practices for cleaning and exploring data with LLMs
- Publish APIs, docs, apps, and more with the Posit Publisher extension
This workshop will be a good fit for you as long as you have a basic familiarity with at least one of R or Python. No installation is necessary—Posit will provide a cloud environment equipped with both Positron and the LLMs we will use during class.
Register now
Spots fill up quickly, so don’t wait! Whether you’re joining us in Houston or attending virtually, these workshops offer an incredible opportunity to learn from some of the best minds in data science.
We can’t wait to see you at posit::conf(2026)!
