A photo of Ed Watson

Mr. Edward Watson

BSc (hons) Psychology and AI; PGdip Technology enhanced learning, innovation and change

Work Address

Innrain 15, 6020 Innsbruck
Office hours: any time by appointment.
please register by email
Mail: Edward.watson@uibk.ac.at
Telephone: +43 512 507 23582

Background

Ed has been working in the area of DevOps since 2021 and more recently taken responsibility for ML operations for the department. He maintains small experimental GPU clusters for the department and leads a small number of development projects. The maintained clusters are used in various research projects; either as large dataset processors (Modin + Ray), data science workbook environments (JupyterLab, Code-server), or local AI runners (Ollama, Llama-server, LM studio).

Ed also works as a Devops engineer and media developer in the department of Digital media and technology (DMLT) for the central IT service of the university. There, he leads the team responsible for the development and operations of the university media portal (Opencast, Tobira) and it's related services. He also engages in external projects on topics to better integrate modern technology developments into teaching.   

In addition, Ed is an appointed external lecturer on AI Media and AI agentics for the UIBK Digital Science Center course. He also runs customised workshops involving AI agentics and spec-driven development for academic and administration staff on the themes of AI uplift and accessible automation pipeline development.

Workshop topics 

1. Human-First AI Automation: A Co-Design Framework for Pharmacy Teams

  • Workshop on collaborative co-design methods to identify automation opportunities in clinical pharmacy workflows
  • Based on human-first methodologies from a 2026 scoping review carried out on the UPLIFT project
  • Focus on ensuring staff engagement, trust, and upskilling with AI agentic tools, rather than displacement
  • Demonstrates a practical toolkit for pharmacy leaders introducing AI into their teams (Skills libraries)
  • Demonstrates a co-design process with facilitator staff of workflow development 
  • Use cases: automating medication reconciliation documentation, prescription processing workflows, clinical trial administration, and regulatory reporting with ensuring GDPR compliance
  • Delivery: 1hr presentation with software demos

2. AI-Powered Research Assistance for Clinical Pharmacy: Building Agentic Pipelines

  • Demonstrates how agentic AI pipelines can assist clinical pharmacy researchers in systematic literature review and data synthesis (e.g., delirium-focused research)
  • Practical demonstration of open vs. closed data researcher assistance tools
  • Implications for evidence-based pharmacy practice and pharmacovigilance research
  • Delivery: 1hr Group Demo presentation AND/OR 1hr participation-based with BYOD laptop and provided n8n/flowise software. 

3. Local AI Models for Pharmaceutical Research: A Practical Guide to Privacy-Compliant AI Infrastructure

  • Hands-on workshop with locally-hosted AI tools for drug interaction queries, formulary analysis, or clinical case summarisation
  • Demonstrates how clinical pharmacy departments can deploy local large language models without sending sensitive patient or research data to external cloud services
  • GDPR/data protection considerations when handling pharmaceutical and patient data
  • Setting up reproducible data science workbook environments for pharmacoepidemiology and outcomes research
  • Processing large-scale pharmaceutical datasets (e.g., prescription databases, adverse event registries)
  • Practical session on collaborative research computing for pharmacy academics
  • Delivery: 1hr Group Demo presentation AND/OR 1hr participation-based with BYOD laptop and provided JupyterLab/Google Collab/Openwebui software
Current projects
  1. DELSTAR 2: Delirium-focused systematic agentic researcher aid. An agent-pipeline for open and closed data researcher assistance.
  2. UPLIFT:  University Personnel Leverage Intelligent Flow Tools.  The UPLIFT project aims to focus on the introduction of intelligent workflow tooling (shortened to just flow tooling) automation for administrative staff at the University of Innsbruck. The project adopts a human-first approach, where project staff collaborate directly with administrative workers to co-design time-saving automations using flow tooling and, where suitable, GDPR-compliant AI services. (https://www.uibk.ac.at/en/pharmazie/klinische-pharmazie/research/uplift-project)
  3. COMPASS: Competency assessment and simulation. An initiative to develop a physical hardware and accompanying software platform to capture student performances of competency in simulated professional environments, for training and human-led assessment.  
  4. UIBK-wide JupyterLab courseware with AI assistance. An initiative to develop a widely applicable, self-service JupyterLab course in an LMS lab environment equipped with the Jupyterlab-AI agent. The project is working along side the UIBK AI landscape team to use local sovereign AI as the LLM source. 
Research publications
  1. Spagl. KT., Watson EW, Weidmann AE. Evaluating a Customised Large Language Model (DELSTAR) and its Ability to Address Medication-Related Questions Associated with Delirium. International Journal of Clinical Pharmacy, 2024 https://link.springer.com/article/10.1007/s11096-025-01900-8
  2. Weidmann AE, Watson EW. Novel opportunities for clinical pharmacy research: Development of a machine learning model to identify medication related causes of delirium in different patient groups. International Journal of Clinical Pharmacy. 2024. 46:992-995. https://doi.org/10.1007/s11096-024-01707-z
  3. McFadyen, M.C.E.; Watson, E.W. Exploring Just-in-Time Teaching 3D Development as a Tool for Enhancing Knowledge and Understanding. Pharmacy 2013, 1, 269-281. https://doi.org/10.3390/pharmacy1020269
Nach oben scrollen