Back to Slides

AI4RA Workshop | REACH 2026

Foundational Module: Framing AI4RA as an Intersection Between Data Science and Artificial Intelligence

Facilitated by Nathan Wiggins

Module structure

Foundations Module Preview

Part 1

AI4RA Introduction

10 min

Part 2

Data Science and AI Intersection

10 min

Part 3

Vandalizer Workflow Demo

10 min

Part 4

Vandalizer Q&A

5 min

Part 5

Evolving Frameworks and Mental Models

10 min

Part 1 · AI4RA Introduction

1. Develop open-source data models and workflows

2. Create trustworthy AI-powered tools

3. Implement and assess the impact of these tools at partner institutions

Part 1 · AI4RA Introduction

🎯 Accuracy

Evidence-based answers grounded in validated data sources.

🔁 Reproducibility

Repeatable workflows with transparent prompts, data, and methods.

🧩 Flexibility

Support for multiple institutional contexts, teams, and use cases.

🛡️ Security

Policy-aligned controls for access, privacy, and operational risk.

Part 1 · AI4RA Introduction

Community of Practice Ecosystem

Vandalizer

AI workflow interface for structured analysis and reproducible outputs.

Vandalizer logo

Data Lake

Shared institutional data foundation to support trusted analytics at scale.

Other Projects

Additional initiatives that extend and apply AI4RA methods to other core research needs.

Part 1 · AI4RA Introduction

AI4RA: The Bridge to Research Analytics

Harnessing Data Science and Artificial Intelligence to:

  • Eliminate barriers that prevent institutions from performing meaningful strategic analysis
  • Manage growing expectations among stake holders and institutional leadership
  • Prioritize consistent data governance and management principles

Part 2 · Data Science and AI Intersection

FAIR Principles as Operational Guardrails

F · Findable

Use persistent identifiers and searchable metadata.

A · Accessible

Provide clear retrieval pathways with governed permissions.

I · Interoperable

Adopt shared formats and vocabularies across systems.

R · Reusable

Document provenance, context, and usage constraints.

FAIR source citation: Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J. J., et al. (2016). “The FAIR Guiding Principles for scientific data management and stewardship.” Scientific Data, 3, 160018. https://doi.org/10.1038/sdata.2016.18

Part 2 · Data Science and AI Intersection

The Intersection of Data Science and Artificial Intelligence

Shift 1

Proactive Data Science

A Shift in Our Reach

Agentic coding and development tools make building new systems attainable in a short period of time.

Traditional data science backgrounds are well-aligned with advances in artificial intelligence.

Endless opportunities and development trees can lead to “Brain Fry.”

Brain Fry citation: Harvard Business Review. (March 2026). “When Using AI Leads to Brain Fry.” https://hbr.org/2026/03/when-using-ai-leads-to-brain-fry

Part 2 · Data Science and AI Intersection

The Intersection of Data Science and Artificial Intelligence

Shift 2

Reactive Data Science

A Shift in Stakeholder Values

Stakeholders now expect faster, more effective solutions powered by AI.

Perceived competition and fear of getting left behind drives initiatives.

Less appeal to dashboards, more appeal to chatbot interfaces.

Part 3 · Vandalizer

  • Built by AI4RA as a practical AI environment for research administration and analytics work.
  • Designed for trustworthy outputs by pairing model responses with clear workflow context and source material.
  • Created to move teams from ad hoc prompting toward repeatable, institution-ready AI practices.

Part 3 · Vandalizer

Workflows are the crowning feature: run the same workflow reliably and repeatedly on unique examples in the same context.

Document-based interface: ground work in real artifacts instead of detached chat alone.

Chat is increasingly prioritized: interaction patterns now reflect how users naturally think and ask questions.

Vandalizer Demo

Vandalizer Questions

Part 5 · Frameworks and Models

Data Science Tasks vs. AI Tasks: Then and Now

BEFORE

Careful distinction between tasks that are well-suited for “data science work” or “AI work.”

AFTER

Agentic workflows call specialized tools that are tailored to the target task

TRUE PRINCIPLE

Task fit matters for optimized results from artificial intelligence

Part 5 · Frameworks and Models

From Prompt Engineering to Intent Engineering

BEFORE

Significant emphasis placed on crafting effective prompts to optimize the output

AFTER

Better processes reduce the need for perfecting prompts with large, mainstream models

TRUE PRINCIPLE

Prompt design helps users get output that aligns with their intent

Part 5 · Frameworks and Models

Amid Constant Change, Data Foundations Stay Essential

✅ Quality

Validated, timely, and complete records.

⚖️ Governance

Clear ownership, policy alignment, and accountability.

🔐 Security

Access controls, monitoring, and incident response plans.

📝 Documentation

Lineage, assumptions, and known limitations.

Bridge forward

Module 2 builds governance depth

Continue with data governance and context engineering to convert this shared framing into concrete operational safeguards.

Open Module 2 facilitator guide