Artificial intelligence is becoming increasingly embedded in how researchers conduct scientific work and how educators design, deliver, and support learning. Researchers are using AI to explore literature, write and debug code, analyze data, generate hypotheses, automate repetitive workflows, and communicate scientific results. At the same time, educators are experimenting with AI-assisted tutoring, assessment, course development, feedback, and personalized learning experiences.

This session explores how AI can be used practically across scientific research and education while also examining the limitations and risks that come with these tools. We will discuss where AI can genuinely improve productivity and discovery, where human expertise remains essential, and how issues such as reliability, reproducibility, privacy, and responsible use should be considered. Practical examples will highlight how researchers, faculty, and technical teams can incorporate AI into existing workflows without treating it as a replacement for scientific judgment or instructional expertise.

Who Should Attend Faculty, researchers, graduate students, research software engineers, educators, instructional staff, data scientists, and technical professionals interested in applying AI within research or teaching environments. The session is especially relevant for those exploring how generative AI and emerging AI tools can support scientific and educational workflows.

Topics

AI-assisted scientific discovery and research workflows Literature review, synthesis, and knowledge exploration AI for coding, debugging, and scientific software development Data analysis, modeling, and scientific computing Generative AI for teaching, tutoring, and learning support AI-assisted course and educational content development Reliability, reproducibility, privacy, and responsible use Practical applications across scientific and academic disciplines

Level Beginner to Intermediate. No specialized AI background is required, although familiarity with common AI or generative AI tools will be helpful.

🔗 Register now: LINK

Retrieval-Augmented Generation (RAG) has become the standard approach for grounding LLMs in external, up-to-date, or proprietary knowledge — but building a RAG system that performs well in production requires far more than a basic retrieve-and-generate pipeline. This session goes beyond RAG fundamentals to cover advanced retrieval strategies, vector database architecture, and the design tradeoffs that determine whether a RAG system is accurate, fast, and scalable. Participants will explore chunking strategies, embedding model selection, hybrid search, re-ranking, and how different vector databases handle indexing and retrieval at scale — all through conceptual explanation, architecture diagrams, and real-world case studies.

🔗 Register now: LINK

Description: The Model Context Protocol (MCP) provides a standardized way for AI applications and agents to connect with external tools, data sources, and services. This session will introduce the motivation behind MCP, its core concepts and architecture, and how it can be used to extend AI applications beyond simple text generation. We will also discuss where MCP fits within the broader ecosystem of AI agents, APIs, and retrieval-based applications.

Who should attend

Students, researchers, and developers interested in AI agents, LLM applications, or connecting language models to external tools and data.

What you'll learn

Participants will learn what MCP is, why it was developed, its main components, and how it can be used to connect AI applications with tools and external resources.

Level

Introductory to intermediate. Some familiarity with LLMs or APIs may be helpful but is not required. The session will primarily be lecture-based, with examples and demonstrations as appropriate.

Tags: AI 🔗 Register now: LINK

As large language models grow in size and capability, deploying and customizing them efficiently has become a critical skill for AI practitioners. This workshop demystifies parameter-efficient finetuning (PEFT) and model compression techniques that make it possible to adapt and run LLMs without massive compute budgets. Participants will explore the theory and intuition behind LoRA and QLoRA, understand how quantization formats like GGUF enable local and edge deployment, and walk through real-world architecture and design tradeoffs.

Who Should Attend

ML/AI engineers and data scientists who want to understand model customization beyond prompting Software engineers building AI-powered products who need to reason about deployment cost and tradeoffs Researchers or students exploring efficient training methods for constrained hardware Anyone curious about how open-source LLMs are finetuned and compressed for local or edge use

Prerequisites: Basic familiarity with how neural networks and transformers work.

🔗 Register now: LINK

Large language models can produce responses that appear useful, but evaluating their quality in a consistent and repeatable way can be challenging. This session will introduce practical approaches for evaluating LLMs and LLM-powered applications, including common metrics, task-specific evaluation, rubric-based scoring, and LLM-as-a-judge approaches. The session will also discuss how evaluations can help compare models, prompts, and application changes over time.

Who should attend

Students, researchers, and developers who use or build applications with large language models and want to better understand how to measure their performance and reliability.

What you'll learn

Participants will learn the fundamentals of LLM evaluation, common evaluation approaches and metrics, and how evaluations can be used to systematically compare and improve LLM-based systems.

Level

Introductory to intermediate. Basic familiarity with large language models is helpful but not required. The session will primarily be lecture-based, with examples and demonstrations as appropriate.

🔗 Register now: LINK

Description: The AI landscape is evolving at an extraordinary pace, with new models, tools, frameworks, and deployment strategies emerging across both research and industry. AI Ecosystem 2026 provides a practical overview of the most important developments shaping the field and helps participants separate meaningful advances from short-lived trends. Rather than focusing only on individual model releases, this session looks at the broader ecosystem, including reasoning models, multimodal systems, open-source AI, agentic workflows, model infrastructure, and the growing role of AI in scientific and enterprise environments.

We will also examine how the AI stack is changing, from foundation models and inference platforms to orchestration tools and emerging standards. The goal is to give attendees a clearer understanding of where the field is heading, which technologies are becoming more important, and what organizations and researchers should be paying attention to as they plan future AI projects.

Who Should Attend Researchers, developers, data scientists, research software engineers, technical leaders, faculty, and professionals who want a current understanding of the rapidly changing AI ecosystem. This session is especially useful for those evaluating new AI technologies, designing AI-enabled workflows, or deciding which tools and trends are worth investing time in.

Topics

Major AI trends shaping 2026 Reasoning models and emerging model capabilities Multimodal AI across text, image, audio, and video Open-source versus proprietary AI ecosystems AI agents and agentic workflows Model orchestration, tooling, and development frameworks Inference, deployment, and infrastructure trends What matters most for research, enterprise, and scientific computing

Level Intermediate. Attendees should have a basic familiarity with AI and generative AI concepts, but deep technical expertise is not required.

Tags: AI 🔗 Register now: LINK

⏰ Time: 3:00-4:30

💻 Location: Virtual (Register at https://events.teams.microsoft.com/event/e66d5247-6b72-45bb-9bf6-78839f275799@4130bd39-7c53-419c-b1e5-8758d6d63f21)

🏫 Instructor: Jacob Verburgt

Who Should Attend

People who are starting to use High Performance Computing systems and want to learn how to use them. There is no prior knowledge/experience requirement to attend these workshops.

What You’ll Learn

By the end of the four 90 minute sessions, you’ll feel confident in your ability to navigate HPC systems and use them to run your workflow. You will learn basic Unix commands as well as HPC-specific ones. By the end, you should be equipped to create scripts and run applications on Purdue's supercomputing clusters.  

Week 1 - Tuesday September 8th, 2026 - Unix basics

Week 2 - Monday, September 14th, 2026 - HPC basics

Week 3 - Monday, September 21st, 2026 - Advanced Unix

Week 4 - Monday, September 28th, 2026 - Advanced HPC  

Level

Beginner - No experience needed

🔗 Register now: LINK

Faculty, staff, and students are invited to present or attend!

First event: July 17, 2026 @ 10am EST

Speakers:

Karolina Willicott – Postdoctoral Associate Florida International University
"Utilizing HPC to study Parkinson's disease modeled in worms"
Karolina finished her PhD at the University of Alabama in 2024. There, she developed substantial expertise in nematode genetics focusing on neurodegenerative disease, at the same time diving deeply into bioinformatics and transcriptomic analyses. Currently, her postdoctoral position at Florida International University includes assembling novel nematode genomes from around the world.

Sanae Moummi - Chemistry researcher with a strong interest in electrochemistry and computational organic chemistry
Her presentation will focus on the development of a solubility prediction model using merged experimental datasets, molecular featurization, machine learning, and graph-based modeling to support faster and more informed material screening.

Sanaae is passionate about applying creativity and design thinking to problem framing, with a focus on human-centered approaches and rapid prototyping. Her most recent work involved designing and testing organic electrode materials for sodium-ion batteries. A major challenge in that project was the solubility of these organic electrodes in the electrolyte solvent. That experience motivated me to develop a solubility prediction model that can help estimate performance earlier in the process, reducing time spent on synthesis and experimental testing.

To attend, please register here: REGISTRATION

Date: Wednesday, July 15, 2026
Time: 1-5 PM (Eastern Time)
Location: Online (Zoom details shared upon registration)
Presenters: Bithi De (RCAC), Cong Gian (Indiana University), and Treven Knight (Mississippi State University)
Registration: Please register by July 6, 2026—REGISTER HERE

You are invited to join the CI-PIVOT (CyberInfrastructure Professionals InnoVating and brOadening the adoption of advanced Technologies) team for a virtual, hands-on NAIRR training workshop focused on practical high-performance computing and AI-enabled scientific workflows.

Event Overview

The workshop will use Purdue’s Anvil system as the primary instructional platform and will introduce participants to the process of preparing, running, comparing, and porting scientific workflows across HPC environments. The training will combine cluster onboarding, MPI-aware HPC usage, comparative Weather Research and Forecasting (WRF) and Pangu-Weather demonstrations, and portable container/Conda-based AI workflow design. For the full schedule, participant requirements and pre-requisites, and registration, see: https://conggian.github.io/WRF_Training_2026/

What You’ll Learn

Anvil-specific HPC onboarding: Open OnDemand, filesystems, Slurm, modules, and a light introduction to MPI as the model for distributed-memory execution on clusters Containerization and AI sidecar workflows: Apptainer definition files and images, build–move–run workflows across systems, and the host-versus-container boundary Conda as a complementary layer for AI tooling and GPU-enabled workflows.

Who Should Attend?

This session is ideal for:

Graduate students, postdoctoral researchers, faculty, research software engineers, and research support personnel who seek a guided introduction to how scientific applications are prepared, executed, compared, and made portable across HPC systems.

No prior experience with Anvil, WRF, Pangu-Weather, Apptainer, or containerization is required. Participants with interests beyond weather modeling are welcome, as the workflow lessons are designed to generalize across computational science domains.

Prerequisites:

• Basic familiarity with Linux command-line navigation
• Basic familiarity with Python is helpful for the AI/Conda portion, but it is not required.
• Prior HPC or Slurm experience is helpful but not required.

Hands-On Participation

To participate in the hands-on components, you must have an ACCESS ID and ensure it has been added to the appropriate project or workshop allocation on Purdue’s Anvil cluster before the session. We recommend testing your login access to Anvil on Open OnDemand in advance at least two weeks in advance.

Registration Reminder

Be sure to register by July 6, 2026 to ensure access to the training environment and resources. Participants should register using their institutional email address rather than a personal email address.

REGISTER HERE

⏰ Time: 3:00-4:30

💻 Location: Virtual (Register at https://events.teams.microsoft.com/event/ab947754-f74a-417b-8678-557518bee6bb@4130bd39-7c53-419c-b1e5-8758d6d63f21 )

🏫 Instructor: Jacob Verburgt

Who Should Attend

People who are starting to use High Performance Computing systems and want to learn how to use them. There is no prior knowledge/experience requirement to attend these workshops.

What You’ll Learn

By the end of the four 90 minute sessions, you’ll feel confident in your ability to navigate HPC systems and use them to run your workflow. You will learn basic Unix commands as well as HPC-specific ones. By the end, you should be equipped to create scripts and run applications on Purdue's supercomputing clusters.   Week 1 - Friday, July 10th, 2026 - Unix basics Week 2 - Friday, July 17th, 2026 - HPC basics Week 3 - Friday, July 24th, 2026 - Advanced Unix Week 4 - Friday, July 31st, 2026 - Advanced HPC

Level

Beginner - No experience needed

🔗 Register now: https://events.teams.microsoft.com/event/ab947754-f74a-417b-8678-557518bee6bb@4130bd39-7c53-419c-b1e5-8758d6d63f21

Registration link: https://purdue.ca1.qualtrics.com/jfe/form/SV_572GkrN6SSeYJSe

Registration link: https://purdue.ca1.qualtrics.com/jfe/form/SV_5cJvPnJax6JjLpk

Registration Link: https://purdue.ca1.qualtrics.com/jfe/form/SV_af1VNhe7NurYvpY

Registration link: https://purdue.ca1.qualtrics.com/jfe/form/SV_af1VNhe7NurYvpY

Registration Link: https://purdue.ca1.qualtrics.com/jfe/form/SV_9HqQtOK6kw9FGf4

About the Camp Your student will step into the role of Mission Control during a simulated Mars mission in crisis! Through three interconnected, story-driven STEM challenges, campers will investigate real problems, design creative solutions, and collaborate with peers - all in a race to save the mission.

This is a full-day, hands-on experience that brings science, technology, engineering, and math to life in a way no classroom can.

Three One-Day Sessions Available

• Session 1 - June 15, 2026

• Session 2 - June 16, 2026

• Session 3 - June 17, 2026

Each session runs from 9:00 AM – 4:00 PM. Snacks and lunch are included.

COST: $50 per camper | Financial aid may be available - please inquire.

Bonus Tours As part of the camp experience, students will also enjoy a guided tour of Purdue's Research Computing Center, including:

• The Data Center - home to Purdue's powerful supercomputing systems

• The Envision Center - where researchers use virtual and augmented reality for cutting-edge data visualization

This behind-the-scenes look at real research technology is a unique opportunity students won't find anywhere else!

Spaces are limited - register today to secure your student's spot!

REGISTER HERE

Registration link: https://purdue.ca1.qualtrics.com/jfe/form/SV_cNrr2Kpb2tS9ifQ

Registration Link: https://purdue.ca1.qualtrics.com/jfe/form/SV_88lX5O7hAfLyqzQ

This session introduces the foundations of modern AI and explains how agentic workflows are built. We will explore how foundation models serve as reasoning engines, how tools and APIs extend their capabilities, how memory enables context retention, and how control loops orchestrate planning and execution. Participants will gain both a conceptual clarity and practical insight into how agentic systems are engineered in real-world environments.

Register Here: https://purdue.ca1.qualtrics.com/jfe/form/SV_038TucaSL7hZvJI

Register here: https://purdue.ca1.qualtrics.com/jfe/form/SV_8uz1ktjhkSyk7dk