Welcome to the SAS Hackathon Bootcamp 2026 experience. In this session you will have the choice of one of four use cases that all cover the full Data and AI Life Cycle.
A description of each use case and its associated industry can be found below. Choose the once you are most interested in or that most closely matches your companies profile. While each use case is different they will all follow the same principals, utilize similar techniques and explain SAS Viya technology along the way.
All data used for these use cases is fictional.
ShopEase, a fictional online retail platform, is losing customers at a 12 % monthly churn rate. The goal is to predict which customers will leave and why — then act before they do.
PremierBank, a regional bank with $2.1 B in assets, faces an 8.5 % loan default rate — well above the 5.2 % industry average — costing an estimated $12.8 M per year. The model must predict high-risk applicants and satisfy fair-lending regulations.
Metro City (pop. 850,000) receives 15,000 service requests per month. Today those requests are triaged manually, causing 40 % response-time variance between districts and longer waits in underserved areas. The model must predict urgency and ensure equitable outcomes.
MedCare Health System (12 hospitals, 2 M+ patients/year) has an 18.2 % 30-day readmission rate — above the national 15.5 % benchmark — costing $12.4 M in annual CMS penalties alone. Predicting which patients are at risk before discharge can save lives and millions.
While each use case is focused on a particular industry while going through them we will be hitting the following five topics:
- Synthetic Data enables us to generate more data or data that can be used in non-production systems. For synthetic data generation we will be using SAS Data Maker.
- Developer Experience and Open-Source enable us to code in different language. For this we are going to make use of SAS Viya Workbench so that you can decide if you want to code in Python, R or SAS.
- Trustworthy AI is the foundation of running processes in production that are touching people and have impact on them. Being able to explain model outputs, understand fairness and mask data to ensure data privacy.
- Copilots are an essential piece to get more out of your existing software faster. With the SAS Viya Copilot we will explore some of its capabilities and see how it can help us build.
- Agentic AI is the way to scale agents in your organization to increase productivity. Within our use case we will be focusing on the power of agentic decisioning to automate processes.
Each use case is separated into five steps, where each one of the steps is task to complete. The steps usually follow each other and build on each other, but for this bootcamp we have created them all in such a way that you can complete them in any order that suits your interests. If you are unsure where to start we recommend the order described below, but it is not required.
| Steps | Purpose |
|---|---|
| 1. Ask & Access | Understand the business problem, identify data sources, form initial hypotheses |
| 2. Prepare | Collect, clean, transform, and organize raw data for analysis |
| 3. Explore | Explore variable relationships, select analytical methods, create datasets |
| 4. Model | Build and run analytical models |
| 5. Deploy & Act | Deploy the solution into production, monitor performance. Visualize findings through dashboards, charts, and reports for stakeholders |
The lifecycle is iterative — teams often move backward and forward between the described steps as new insights emerge. During this bootcamp we will not have the time to do that, but if you are interested in having a follow up on that topic specifically please reach out by talking to your onsite SAS Hackathon Bootcamp mentor directly.
Access to the SAS Viya environment, SAS Viya Workbench and SAS Data Maker will be provided onsite at the SAS Hackathon Bootcamp event. Please do not forget to bring your own laptop.