Enabling Reproducible AI Workflows for the NAIRR Ecosystem
NAIRR 2026 Tutorial Schedule
Slides for this tutorial are here.
| Time | Duration | Description |
|---|---|---|
| 9:00-10:00 AM | 60 min | (Lecture) AI/ML, the National CyberInfrastructure Ecosystem and Tapis |
| 10-10:30 AM | 30 min | (Hands-on) Initial Steps With TapisUI |
| 10:30-11 AM | 30 min | Coffee Break |
| 11-11:45 AM | 45 min | (Lecture) Classes of AI Models, 3rd Party Model Registries, Flexserv |
| 11:45-12:30 PM | 45 min | (Hands-on) Executing Large Models on Vista Via Tapis and Flexserv |
| 12:30-1:30 PM | 60 min | Lunch |
| 1:30-2:00 PM | 30 min | (Lecture) Computer Vision Models in Scientific Applications |
| 2:00-2:15 PM | 15 min | (Hands-on) Launching Jupyter Notebook |
| 2:15-2:45 PM | 30 min | (Lecture) Prompt-Engineering: Best Practices (Hands-on) Generating Code for a Vision Task and Executing in Jupyter |
| 2:45-3:00 PM | 30 min | (Lecture + Hands-on) Fine-tuning the Vision Model with Ultralytics and Tapis |
| 3:00-3:30 PM | 30 min | Coffee Break |
| 3:30-4:15 PM | 45 min | (Hands-on) Re-evaluating Using best.pt within Jupyter |
| 4:15-5:00 PM | 45 min | Additional Use Cases and Discussion |
The resources you will be using today are provided by funding from the National Science Foundation.
Intro to Tapis
Links & Documentation
The tutorial covers TapisUI in the public tenant: public.tapis.io
Live Documentation of OpenAPI V3 Specs for all Tapis Services
Full documentation and guides for using Tapis
TapisUI specific documentation page
