Posit at Phuse EU 2026

Modernize Your Clinical Workflow. The future of clinical reporting is open, reproducible, and AI-assisted. Whether you're moving to R and Python-based submissions, scaling Shiny apps for safety monitoring, or testing where GenAI fits in drug development, Posit can help. 

100% of the top 20 pharmaceutical companies use Posit. Join us at booth 43 in Glasgow, November 15th to 18th, to see how clinical teams are building the next generation of clinical data science.
 

Book time with Posit

WORKSHOP & TALKS

Join Our Sessions

Join us throughout the conference for deep dives into the future of clinical programming.

Sunday, November 15 

14:30 - 16:00 GMT | WORKSHOP: Agentic AI in Clinical Programming: Separating Hype from Production-Ready Practice with R/Python | Phil Bowsher 

AI-assisted coding has moved from demonstration to deployment — but what works in practical, regulated environments? This workshop cuts through the noise to show what is ready for use today. We explore the questions teams are actively asking: how to achieve consistent, traceable outputs from AI tools, how to keep programmers in the loop, and which governance patterns are emerging across the industry.

This workshop demonstrates implementation patterns through hands-on exercises. Participants will work with Positron (Posit’s polyglot IDE with native AI capabilities), Model Context Protocol (MCP) servers, and open-source tools (ellmer, ragnar, shinychat) to build LLM-assisted ADaM spec-to-code workflows with human-in-the-loop validation. Participants will leave with practical approaches, reusable patterns, and ideas they can take back to their teams.

Key Takeaways:

• Technical trade-offs: Evaluating deterministic versus probabilistic code generation for CDISC programming and trial integrity
• Compliance-friendly workflows: Building AI pipelines that support audit readiness, access controls, and traceability within enterprise SCEs
• Hands-on automation: Developing governed AI agents that convert complex study specifications into standards-aligned R code

Statistical programmers and data scientists will gain hands-on experience with practical AI tooling workflows, structured prompt engineering, and Positron’s native IDE features. Biometrics leads and IT architects will learn operational strategies for human-in-the-loop oversight and secure pathways for progressing AI deployments from pilot to production workloads.

Monday, November 16

16:00 - 16:30 GMT | TALK: Open-Source AI Pit of Success: Governance within the ACE (AI Statistical Environment) | Phil Bowsher & Joe Cheng

This presentation explores the transition from Statistical Computing Environments (SCE) to the ACE (AI Statistical Environment) where AI availability and governance is integrated for users. We demonstrate how organizations implement a "AI Pit of Success" for open-source users via Posit Integrations, ensuring AI is open-source aware and ready. Attendees will learn how these tools integrate within an ACE infrastructure, providing users with the Posit Assistant for AI enablement. The session highlights the role of the "Clinical AI Fabric" in providing a secure, VPC-internal orchestration layer with Amazon Bedrock (or similar) in the ACE. Finally, we showcase how this governed ACE approach enables human-in-the-loop reviews and agentic workflows, providing a foundation for AI code-first of clinical reporting.

Tuesday, November 17 

12:00 - 12:30 GMT | TALK: Agentic QC for R and Python in Positron | Phil Bowsher & Adrian Waddell

This presentation discusses "Agentic QC," a shift where AI agents act as validation partners within the Positron IDE. We explore how to enhance the manual challenge of double programming by orchestrating Agents that support deriving, SDTM and ADaM datasets and TFL outputs as the primary programmer works. The session demonstrates how this ACE (AI Statistical Environment) capability utilizes Claude Code and the Posit Assistant to cross-reference code against company standards and Pharmaverse best practices. Attendees will learn how to support GxP-level rigor through an "automated check" that identifies challenges  before formal validation begins. Finally, we showcase how Agentic QC supports data integrity while aiding the time-to-submission for late-stage clinical reporting.

Essential Resources for Clinical Programmers

Curated insights for the PHUSE community.

• Schedule an AI Workshop for your team
• Schedule 1:1 time to speak with Phil Bowsher
• Posit’s Open Source Clinical Reporting Playbook
• Posit’s work in AI
• Get started with our new IDE: Positron