10 Years of Data Science Tools... and What Happens Next
In this talk, I'll reflect on a decade of work on RStudio and the principles of tool-building that have led it to become the standard data science environment for R. We'll talk about how those same principles have guided the development of Positron, a new data science environment from Posit, and how you can apply them to your own tool-building work.
AI Coding Assistants: Hype, Help, or Hindrance?
AI coding assistants like ChatGPT, GitHub Copilot, and Codeium promise to revolutionize our coding workflows—but how useful are they in practice? Are they our new overlords here to take our jobs? Or just a passing gimmick? I think the reality lies somewhere in between, and that understanding these tools is key to staying relevant in today’s rapidly evolving data science ecosystem.
In this talk, I'll show how I’ve used these AI tools in RStudio, Positron, and VS Code to speed up both my advanced R workflows as well as my learning experience as an intermediate Python programmer, providing examples, pitfalls, and best practices.
AI missteps as stepping stones: Opportunities gained when your LLM coding assistant gets it wrong
LLM coding assistants have become a valuable companion for learning and productivity in data science. (Hey Siri! Import this csv!) While their ability to generate code and explanations is impressive, I have found more value and personal growth from the mistakes they make. This talk focuses on embracing coding assistants as imperfect companions and succeeding when they fail. I'll share insights from using these assistants to facilitate my transition to Python, highlighting the pitfalls of accepting their recommendations without question. Through real examples where LLMs fell short, I'll demonstrate how these challenges provided frameworks for problem-solving and led to a deeper understanding of data science tools and methodologies.
Air - A blazingly fast R code formatter
In Python, Rust, Go, and many other languages, code formatters are widely loved. They run on every save, on every pull request, and in git pre-commit hooks to ensure code consistently looks its best at all times.
In this talk, you'll learn about Air, a new R code formatter. Air is extremely fast, capable of formatting individual files so fast that you'll question if its even running, and of formatting entire projects in under a second. Air integrates directly with your favorite IDEs, like Positron, RStudio, and VS Code, and is available on the command line, making it easy to standardize on one tool even for teams using various IDEs.
Once you start using Air, you'll never worry about code style ever again!
AMA with Rich Iannone
Get ready for an AMA with Rich Iannone, creator of the {gt} and {pointblank} packages and all-around great human. Let’s chat with Rich about package dev! But, hey, it’s an AMA (ask-me-anything) - so ask whatever you want!
AskRADS: An AI Recommendation Agent for Maximizing the ROI of Data Science Collaborations
Blockers to crucial data-driven decisions can often be a challenge. To address this, we established RADS, the Regeneron Analysts and Data Scientists, as a Community of Practice for exchanging strategies on eliminating these obstacles. RADS has grown to nearly 500 members, creating a new challenge: avoiding redundancy and helping non-RADS colleagues find the right experts. To solve this, we developed AskRADS, an AI agent on Posit Connect that provides recommendations based on discussions, experts, and relevant resources. It uses R, Shiny for Python, FastAPI, LangGraph, Neo4j GraphRAG, and MySQL. This talk will cover its architecture, AI search solutions, and optimization techniques.
A tale of one organization: learning how to swim in the ocean of open source
The Canadian Institute for Health Information is diving into open source, transitioning our 300+ analysts and all our analytics code from proprietary software to R and Python. We started 3 years ago, and over that time we have migrated 700,000+ lines of code, navigating waves of change while adapting to new IT environments and workflows. Challenges like technical barriers, workload balancing, and resistance to change were tackled through training, mentorship, and executive support. More than tools, this shift transforms our culture—turning siloed developers into fearless swimmers. Success requires training, teamwork, and courage. Organizations making a similar leap should start small, invest in training, and stay adaptable to stay afloat.
Beyond the Basics: Expanding Quarto's Capabilities with Lua
Are you familiar with Quarto and eager to push its boundaries? This session is for those ready to explore the power of Lua for customization. Whether you're a novice implementing simple Lua filters or a seasoned developer seeking inspiration for Quarto Extensions, this talk offers valuable insights into Pandoc's and Quarto's Lua features.
We’ll explore Quarto's unique Lua support, including custom AST nodes and helper functions, showcasing how both straightforward and advanced techniques can transform your documents. Through practical examples, you'll gain the confidence to extend Quarto's functionality and unlock new possibilities.
Join us to elevate your Quarto projects and contribute to its growing ecosystem!
Bold indicates negative?
Over one billion people worldwide use spreadsheets to manage and analyse data, often styling cells and their contents to highlight or distinguish values. This formatting is often used to encode additional data (for example, indicating groups with colors) but most data science tools are unaware of data expressed as formatting. This talk summarizes my progress addressing the gap between formatted spreadsheets and the modern data stack. First, by championing data organization best practices, then by bringing spreadsheet contents and their formatting together into R for further analyses (unheadr package), and finally how translating format to data helped me develop tools (forgts package) for converting formatted spreadsheets to gt objects.
Breaking Barriers: Making Ecological Modelling Accessible with EcoCommons
Would koalas prefer to live on the East Coast or West Coast of the US?
Answering complex questions like this requires robust and scalable technology. Enter EcoCommons, a cutting-edge platform built on an R package that’s scientifically rigorous and designed for scale. EcoCommons efficiently runs millions of models to help researchers, governments, and NGOs make data-driven decisions. Whether it's predicting habitat loss, optimising conservation strategies, or advising policy, this platform turns R into a powerhouse for real-world impact—all while building a thriving community of practice. In this talk, we’ll explore the challenges of running R at scale, lessons learned, and how our community is shaping the future of modelling.
Building a Real-Time COVID-19 Surveillance System with R: Lessons from the COVID Symptom Study Sweden
During the COVID-19 pandemic, the COVID Symptom Study Sweden collected over 20 million daily health reports from more than 200K participants. This talk demonstrates how R served as the backbone for transforming this massive dataset into actionable public health insights. I'll showcase our analytics pipeline built entirely in R, from processing raw data to developing an interactive Shiny dashboard for real-time COVID-19 surveillance. The presentation covers predictive modeling for prevalence and hospital admissions estimates and the creation of our 'covidsymptom' R package. Through practical examples, I'll share key learnings about handling large-scale health data and creating data products with R during a public health emergency.
Building Governable ML Models with R
For a model to provide value in production, it must be fit for purpose, deployable, and maintainable over time. We know that R provides a host of tools and packages for building good models, but the language and ecosystem also provide tools to help us build these kinds of maintainable production systems. This talk will present techniques, adapted from software engineering, that provide a stable foundation for building models and writing all the accompanying code that's often needed to train, test, and update models over time. Those attending this talk will learn how, by centering model development on packages, writing tests, creating intuitive S3 methods, and more, we can build modularized, testable code that makes our models easier to monitor and update over time.
Building Multilingual Data Science Teams
For much of my career, I have seen data science teams make the critical decision of deciding whether they are going to be an “R shop” or a “Python shop”. Doing both seemed impossible. I argue that this has changed drastically, as we have built out an effective multilingual data science team at Ketchbrook, thanks to polars/dplyr, gt/great-tables, ggplot2/plotnine, arrow, duckdb, Quarto, etc. I would like to provide a walk through of our journey to developing a multilingual data science team, lessons learned, and best practices.
Building Web Apps and APIs in R with Ambiorix
Ambiorix reimagines web development in R, offering a flexible, autonomy-driven alternative. It gives developers full control over the request-response cycle, while providing essential web features like routing and middleware out of the box—ideal for developing large web applications and APIs. This talk will cover what sets Ambiorix apart, its key advantages for developers, and how it streamlines building web products & services in R. Whether you're an R developer looking to build web apps with more flexibility, or exploring new tools for APIs, this session will show you why Ambiorix is worth considering.
Can I trust that package?
We often forget, surrounded by hex stickers and bad R package puns, that not everyone is as trustworthy as us. This, I suppose, means that when IT asks, “Is this package valid, secure, and trustworthy?” it’s not that unreasonable a question. But this throws up multiple issues. There are thousands of R packages on CRAN, and that doesn’t include the R-universe, Bioconductor, and GitHub. Packages are updated all the time, so how do we keep up? More to the point what does valid, secure and trustworthy even mean?
In this talk, I’ll discuss the litmusverse. A suite of packages for assessing a package risk. Importantly, it’s not one size fits all. Instead, it’s about defining your risk appetite and acting accordingly.
ChalkTalk: Globalizing Data Science Education with AI-generated Videos
We present ChalkTalk, an open-source tool that converts Quarto documents into engaging educational videos with AI-powered voices and avatars. By adding simple text-to-speech (TTS) and text-to-video (TTV) attributes to markdown files, educators can automatically generate multilingual video content while maintaining the reproducibility benefits of Quarto. At The GRAPH Courses, where we've trained over 3,000 learners globally, we are testing out this tool to scale our video content creation. We'll demonstrate its integration with Quarto and present preliminary findings from our A/B testing with students.
Championing modern science workflows to benefit dairy farmers
Dairy research faces data volume and variability challenges. DairyNZ's Modern Science Workflows project addressed this via infrastructure, capability, and business disciplines. Infrastructure: Snowflake cloud data warehouse and Posit Workbench for R. Capability: R data science courses (Uni of Waikato & internal). Business: Meetings, documented code, best practices, and moving towards open science with GitHub at organisational level. Results: We have enabled large dataset projects utilising machine learning which would not have been possible before (e.g., animal sensor data analysis). Training has been well-received. Modern workflows enable reproducibility, and data science skills should become standard scientific training.
Cooking with Data: A Recipe for Data Validation Using Pointblank
Data validation is an exercise that is commonly forgotten. It can be difficult to start the exercise, let alone figure out a holistic approach to cleaning and understanding data. This is a problem that affects us all regardless of experience, expertise, or complexity of work. With Pointblank’s suite of validation functions, you can scan your data, present that scan in a visually appealing and unique report, and then define validation rules at scale in a methodical manner. I want to show why pointblank, above other validation packages, is the best validation approach through its simple yet appealing reporting features and comprehensive validation schema.
Data 911: how Posit can support decision-makers in times of environmental crisis
Over 200 million gallons of mining wastewater was released into Tampa Bay in March of 2021. Concerns about the environmental impacts prompted a multi-agency response to monitor water quality changes in the bay, producing thousands of sample points in need of synthesis and communication to a concerned public. This talk will describe how the Tampa Bay Estuary Program leveraged Posit products to create a data synthesis workflow and Shiny dashboard to inform decision-makers on how, where, and when water quality was affected by this pollution. Our experience navigating this event in real time will be shared with the broader community as a successful example of how Posit products can address environmental crises.
Data-as-a-product: A framework for collaborative data wrangling
Data preparation requires substantial time and subject matter expertise but is often tailored to a single-use deadline rather than encouraging reusable workflows across a team. We developed a framework that acknowledges the time and expertise invested in data preparation and maximizes its value. Our data-as-a-product suite of R packages promotes joint code and data version control, standardizes metadata capture, tracks R package versioning, and encourages best practices such as adherence to functional programming. I'm excited to share my experience onboarding collaborators to this reproducible research framework, highlighting key challenges and lessons learned from advocating for good development practices in a dynamic research environment.
Data Science Hangout ft Matthew Montero, Chief Data Officer at Gen Re
Join us for an in-person Data Science Hangout with Matthew Montero, Chief Data Officer & Head of Enterprise Data and AI Services at Gen Re. New to the Hangout? It's a weekly online gathering for the data science community, running since July 2021. Each week, a different leader helps us kick off an open discussion about leadership, career lessons, workflows, and much more. The questions and conversation are all powered by you! Come hangout with us this afternoon!
Data Science Hangout (Q&A) with Virtual Talk Session Presenters
Talks will be followed by a cozy hour-long Q&A with the virtual speakers hosted by Libby Heeren & Rachael Dempsey. If you’ve ever attended a Data Science Hangout, you’ll love this Q&A! We’ll cover questions asked in Slido and chat with each other in the Discord server.
Death by Dropdown? Engineer Insightful Shiny Apps with Behavioral Science
Opening a Shiny app to endless dropdowns and scattered filters leads to immediate disengagement and feeling like "death by dropdown". This talk introduces the Behavioral Insights Design (BID) Framework, a behavioral science roadmap that transforms overwhelming dashboards into guided user journeys. Using the "Interfaces Need A Smart Vision" mnemonic, we'll explore five stages: Interpret user needs through data storytelling, Notice cognitive friction points, Anticipate user behavior, Structure the application, and Validate & empower users. Drawing from established theories and real-world production examples, attendees will see how applying behavioral science can dramatically reduce the time between "Where do I even start?" and "Aha, now I get it!". You'll leave with practical techniques using the new {bidux} package to help transform user confusion into confident exploration and build dashboards that guide users toward better decisions (and fewer ghosted users).
Demystifying MLOps with Vetiver
MLOps is the process of setting up a Machine Learning lifecycle, including model training, deployment and monitoring. It is a complex topic which brings together an understanding of data processing, modelling and cloud architecture. It is therefore not surprising that many newcomers (myself included) can feel intimidated by the subject. In this talk I will draw on my experience as an organiser of local data science meetups. I will go into how MLOps is often presented within the data science community, how it could be made more accessible to students and beginners, and my current process for teaching MLOps in R and Python using my favourite package, vetiver. In summary: no, you do not have to be an expert in AWS or Azure to get started!
Deploy with Confidence: Strategies to Minimize Failure and Maximize Success (on Posit Connect)
Release day has arrived - you’re eager to share the latest version of your app with users. You deploy it, ready to celebrate…
Then everything breaks. The app won’t load, and the errors make no sense. Despite all your efforts - testing locally, verifying in staging - production still surprises you.
Deployments are tricky, full of hidden challenges that can derail even the best preparations. But with the right strategies, you can reduce risk and increase the likelihood of a successful release
In this talk, we’ll explore deployment techniques such as blue/green deployments and feature flags, showing how they can be applied when working with Posit Connect. You’ll also learn what to do when the unexpected happens.
Design of Everyday Shiny Apps
Donald Norman’s Design of Everyday Things shows how even smart people struggle with bad design—flipping the wrong switch, pushing a door instead of pulling. The same issues exist in Shiny apps, where function often outweighs experience. But as a Shiny developer, you are the door designer, and it’s your responsibility to create user-friendly tools. Not to worry—good design isn’t just for designers. You don’t need an art degree to build intuitive, visually appealing apps. This talk covers composition, color theory, scale, and proportion—practical rules to improve usability. We’ll show real-world examples and code-based strategies to bring these principles to life. Because good design isn’t a luxury—it’s a necessity.
Disposable Shiny Apps
Many data scientists find themselves building Shiny apps for one-off presentations, client meetings, or teaching demonstrations. These "disposable" apps can suffer from overengineering, leading to unnecessary development time and complexity. Or, the apps never get built because the development hill is too high to climb. This talk covers disposable Shiny apps - intentionally minimal applications designed for specific, short-term needs. We'll explore strategies for rapid development, including reusable templates, coding assistants, efficient styling, and design principles that prioritize speed and clarity. This talk will show how this approach can transform a typical week-long development process into a few hours while maintaining polish.
duckplyr: Analyze large data with full dplyr compatibility
The duckplyr package is now stable, version 1.0.0 has been published on CRAN. Learn how to use this package to speed up your existing dplyr codes with little to no changes, and how to work with larger-than-memory data using a syntax that not only feels like dplyr for data frames, but behaves exactly like that.
Election Night Reporting Using R & Quarto
Election night reporting (ENR) is often clunky, outdated, and overpriced. The Idaho Secretary of State’s office leveraged R and Quarto to create a better ENR product for the end user while driving down costs using the open-source software we all know and love. With help from Dr. Andrew Heiss, R was used in every step of the process—from {dbplyr} backend to visualizing the results using {reactable} tables and {leaflet} maps, combining the output into a visually appealing Quarto website. Quarto was the ideal solution due to its scalability, quick deployment, responsive design, and easy navigation. In addition, Dr. Heiss will discuss the advantages of using a {targets} pipeline and creating programmatic code chunks in Quarto.
Elevating Public Health Decision-Making with R Packages
Supporting public health decisions in high-stakes environments requires transparency, reproducibility, and efficiency. Analyzing real-world health data with complex models helps policymakers mitigate infectious disease spread. Structuring projects as R packages provides a consistent framework that enhances organization, integrates documentation, and facilitates collaboration. This approach improves coding practices, ensures reproducibility, and enables seamless sharing of tools—empowering colleagues without the resources to develop their own. This talk will demonstrate how adopting R package structures can enhance workflows and impact without requiring advanced software development skills.
Empowering Learners with WebR, Pyodide, and Quarto
WebR, Pyodide, and Quarto are powerful technologies that let you run code exercises in the Web browser. Because of this, WebR exercises can be integrated into data science lessons within RevealJS slides and Quarto websites. In this talk, I want to emphasize some considerations for using WebR/Pyodide for active learning in the classroom.
Careful exercise design with WebR/Pyodide can make the difference between empowering learners and demotivating them. With our R-Bootcamp and other exercises as examples, I show scaffolding methods for teaching data science concepts gradually, as well as other design considerations. I'll finish up with showing you how to set up WebR up in your slides and websites for your Data Science Learners.
Enemies to lovers: How non-programmers can make sparks fly when using testthat during package development
You’re just a data scientist, self-taught in R, trying to find your way in the world of package development. It’s just a package named testthat, hoping to help a developer make sure their package is operating as intended. You meet. You hate each other. The package seems a little daunting and quite frankly, a little tedious. Surely only “real” programmers use this! The package thinks you’re inept for not immediately putting it to use. As fate would have it, you attend this talk and discover you are indeed compatible. Sparks fly, a package is born. This talk will help less experienced programmers learn about testing, automated workflows, how to write good tests, and why it’s all worth it when it pays off in quality and efficiency. Steamy.
Exploring Datasets in Positron
Inspecting raw data in data frames and tables can be a critical tool in the data preparation, tidying, and feature engineering process. In Positron, we made it a priority to design a modern Data Explorer component that works well for both large and small datasets. In this talk, I will discuss the design of the Data Explorer UI and its backends for Python, R, and DuckDB, and how we made it work smoothly with massive datasets having millions of rows or thousands of columns. Additionally, I will discuss the sorting, filtering, search, and statistical data visualization capabilities that we have added to help make users more productive.
Extending the horizons of R with Rust
Data volumes have skyrocketed for years, outpacing advances in hardware. When R users hit performance bottlenecks, the traditional remedy has been to reach for C++ and include it in their R code using Rccp. In the last few years Rust has emerged as a modern, high-performance alternative for extending R. In this talk, we’ll explore why Rust is a natural fit for data teams—from its robust safety to its concurrency advantages. You’ll also see real-world case studies of how organizations are leveraging Rust and R together to tackle large-scale, compute-intensive challenges. Join us to learn how Rust is expanding the horizons of what’s possible in R, some tips on extending R with Rust, and how your team can benefit from these new possibilities.
Failure (and Mistakes)
In a field driven by precision, the power of failure is often overlooked. This talk digs into the paradoxical benefit of error in data science, drawing on high-profile missteps in data handling and personal anecdotes of falling short. Using examples from errors big and small leading to impacts big and small to the everyday misinterpretation or misuse of data that happens everywhere, we’ll focus on how to get the best out of failure. While some level of error is inevitable in data science, the most resilient and forward-thinking teams realize that errors can drive innovative and creative solutions that may not have been discovered if everything had gone as planned.
From Framework to Function: Integrating AI, Data Science Platforms, and Psychological Safety
Navigating the vast array of AI and data science tools can be daunting. This talk details our journey to address this challenge, starting with an enterprise data architecture and followed by the development an AI framework—a foundational layer of AI services. We advanced by integrating developer-preferred data science platforms, such as Posit Workbench and Posit Connect, alongside Databricks for centralized data governance. Our choices have enabled us to use Posit Workbench and Databricks to perform analyses and reporting with enhanced efficiency and alignment with modern architectures and governance standards. The rapid integration of these services was driven by a focus on psychological safety, a catalyst for enhancing team performance.
From messy to meaningful data: LLM-powered classification in R
Transform messy, unstructured data into meaningful, structured data using ellmer, Posit's R package for interacting with large language models (LLMs). LLMs can quickly convert thousands of texts, PDFs, and images into the data you need for your analysis. This talk demonstrates the practical benefits and addresses the common pitfalls of using LLMs through three examples: classifying images of Iris flowers and text descriptions of disease symptoms and crimes. Learn structured outputs, prompting strategies, model accuracy and confidence measurements, and validation techniques that blend generative AI with traditional ML concepts. Perfect for both AI-curious beginners and experienced data scientists.
From SDKs to Agents: Building with R and Python on Databricks
Databricks offers a rich ecosystem of packages for R and Python developers. This session explores the key tools available—like {ellmer}, {brickster}, and the Databricks SDKs—helping developers build scalable workflows and AI-powered applications. Whether you're working in R or Python, you'll learn how to integrate your IDE, leverage Databricks-hosted LLMs, and build agentic workflows. We’ll cover practical implementations, best practices, and how to scale your projects using Databricks’ robust infrastructure. Whether you're automating workflows or deploying AI-driven solutions, this talk provides the essential toolkit for success.
From Solo to Social: Making Coding a Collaborative Adventure
What if coding were not an exclusive skill but an accessible adventure? Through community-driven, project-based learning, I transform data science education into an inclusive journey. My approach builds trust by meeting participants where they are, using familiar tools and relevant datasets. Through targeted workshops, and collaborative hackathons, I break down complex coding concepts into digestible applications for global audiences.
By reimagining how we teach technical skills, I'm not just delivering education—I'm fostering a movement that makes data science accessible to all. Join me to discover how we can democratize coding and create pathways for everyone to become active members of the data science community.
Get the Latest on Posit's Commercial Products
Join us for an overview of the latest developments across Posit’s commercial product ecosystem. This session will cover Posit Workbench, Package Manager, Connect, Connect Cloud, and our growing portfolio of managed services including Snowflake and beyond. Hear directly from the product managers and engineers who are building these tools, and get insights into what’s coming next.
Attend this session to hear about:
- Audited jobs, Positron Pro sessions, and GenAI in Workbench
- Auth and integrations with RStudio Pro sessions in Package Manager
- An intro to Chronicle for Posit Team
- Building container images in Connect
- Organization plans in Connect Cloud
- A Snowflake Native App offering for Connect and Workbench
- An intro to Posit Team Dedicated
Get your ducks in a row... faster Shiny apps with DuckDB
Our small team tackled sluggish Shiny applications by implementing DuckDB as a cache layer, transforming slow, resource-intensive operations into responsive user experiences. Initially, our application pulled data directly from SQL Server, performing real-time aggregations that resulted in poor response times. By restructuring our pipeline to pre-compute results in DuckDB, we significantly improved performance. This presentation demonstrates how we identified bottlenecks, implemented DuckDB integration, and measured improvements. We'll share practical examples of DuckDB integration with R, discuss trade-offs, and show how this accessible solution can benefit other small teams with limited resources.
Hacking Productivity with LLMs: What Works (and What Doesn’t)
Over the past 18 months, I've used large language models (LLMs) in a series of experiments to improve my productivity as a developer on open source projects in R. I wanted to see if these tools could help with the repetitive, frustrating, or time-consuming parts of development, freeing me up to focus on the interesting problems. The results were mixed: some of these experiments were wildly successful, but others were clear failures.
In this talk, I'll share what worked, what didn't, and what I learned. I'll explore important questions to ask around the kinds of tasks LLMs are good for and where they fall short, how to use LLMs to build lasting skills, and why in some cases, investing in human connection pays off more than AI assistance.
History repeats itself: What the Du Bois Challenge taught me about reproducing visualisations
In 2024, I participated in the Du Bois challenge to recreate W.E.B Du Bois’s iconic 1900s charts on African American sociology. By reproducing the old graphs and their annotations, themes, accessibility and visible contrasts, I levelled up my visualisation skills. I learnt about complexity of design, colour palettes, fonts, styling, and of course appropriate R packages. This experience empowered me beyond the DuBois challenge and used what I learnt to take part in the 30-day chart challenge, Genuary and Tidy Tuesday. In this talk, I will share what I learnt so that you too can more easily become familiar with unfamiliar charts and craft your own visualizations to regale stories.
How I got unstuck with Python
Python as a language is known for being explicit, simple, readable, and beautiful. At the same time, the tooling around using and writing this language has not always made people feel productive and delighted. I know this has been true for me! In this talk, learn about recent improvements in tooling for Python that have finally addressed my own persistent challenges. Posit’s new IDE, Positron, provides a next generation environment for Python data practice, and this new IDE plays nicely with modern language tooling from the Python community. Whether you are Python curious or looking for ways to improve your Python workflows, hear about how I finally got myself unstuck with the most popular programming language in the world.
IDE-ntity Crisis: Choosing the Right Tool for Me
Data practitioners have more IDE choices than ever, but not all are built for the same purpose. JupyterLab prioritizes a notebook-first experience, while VS Code focuses on software engineering with extensible features. Others, like Positron, are designed specifically for data scientists, streamlining workflows and boosting productivity.
Understanding these trade-offs helps you find the best fit. By the end of the session, you'll have a clear framework for choosing the right IDE for your data science and analytics tasks.
Instant Impact: Developing {docorator} to Simplify R Adoption for Teams
Although R supports comprehensive analysis workflows, creating polished, production-ready PDFs directly from R remained a challenge for our pharma teams. With teams facing looming deadlines, our R enablement team swiftly created {docorator}—an open-source R package that transforms R-based tables and figures into production-level PDFs. By adorning results with “decorations” like headers, footers, and page numbers, {docorator} produces seamless, polished documents. Powered by Quarto, it also auto-sizes {gt} tables for user ease. Attendees will learn how {docorator} became the missing piece in GSK’s R workflows and learn how focusing on quick, simple solutions can have a lasting impact.
Is that LLM feature any good?
The ellmer package has enabled R users to build all sorts of powerful LLM-enabled tools. How do you test these features, though? How do you know whether a change to your prompt made any difference, or if a much cheaper model would work just as well for your users? This talk introduces an R port of Inspect, a Python framework for LLM evaluation that has been widely adopted by both LLM developers and tool builders. Attendees will learn about the process of—and importance of—evaluating LLM-enabled apps empirically.
Keeping Data Alive: Persistent Storage Options for Dynamic Cloud Applications
Building cloud-based data applications that evolve over time requires persistent storage solutions. Without a way to record new information or modify datasets, applications like Shiny or Streamlit would reset after each session, losing potentially valuable information. This talk explores persistent storage options ranging from lightweight solutions like Google Sheets to scalable services like Amazon S3 and MotherDuck. We’ll also cover securing connections to these services using Posit Connect Cloud’s secret variable management. By the end, you’ll understand common persistent storage solutions, their trade-offs, and how to choose the best approach for your specific project.
Leveraging LLMs for student feedback in introductory data science courses
A considerable recent challenge for learners and teachers of data science courses is the proliferation of the use of LLM-based tools in generating answers. In this talk, I will introduce an R package that leverages LLMs to produce immediate feedback on student work to motivate them to give it a try themselves first. I will discuss technical details of augmenting models with course materials, backend and user interface decisions, challenges around evaluations that are not done correctly by the LLM, and student feedback from the first set of users. Finally, I will touch on incorporating this tool into low-stakes assessment and ethical considerations for the formal assessment structure of the course relying on LLMs.
Lift Off! Building REST APIs that Fly
Picture the scene: you've successfully deployed your ML model as a plumber API into production. Your company loves it! One team uses the API's predictions as an input to their own ML model. Another team displays the predictions in an internal Shiny app. But once adoption reaches a certain point, your API's performance starts to degrade. What can you do to help your service maintain high performance in the face of high demand? In this talk, we'll show some strategies for taking your API performance to the next level. Using two R packages, {yyjsonr} and {mirai}, we can augment our API with faster JSON processing and better responsiveness through asynchronous computing, allowing our services to do great things at scale at no additional cost.