A Day in the Life of Ramona: Building Trusted Knowledge with SAS® Retrieval Agent Manager
Discover, through the eyes of an EMS education director, how to turn scattered PDFs into a trusted, living knowledge system. In this post and demo video, you’ll see how SAS® Retrieval Agent Manager delivers instant, grounded answers, automates updates, and connects knowledge to real‑world workflows—so teams get the right guidance exactly when it matters.
Setting the Scene
Meet Ramona. She’s the Director of Emergency Medical Services (EMS) Education at a regional healthcare network.
She ensures every Emergency Medical Technician (EMT), paramedic, and dispatcher has the right training.
EMS is the backbone of emergency response. EMTs deliver Basic Life Support (BLS). Paramedics deliver Advanced Life Support (ALS).
Her work touches protocol updates, compliance audits, and training materials.
Her challenge?
Most of her knowledge is trapped in static PDFs: agency guidelines, Standard Operating Procedures (SOPs), research articles. These documents don’t “talk,” so cross‑referencing is slow and manual.
Success?
Success for Ramona is instant, grounded answers with full traceability. She wants to know what was answered and why it was answered.
Video
Watch the demo to see how SAS Retrieval Agent Manager supports Ramona at key moments throughout her day.
(SAS® Retrieval Agent Manager stable version 2026.01. RAG = Retrieval‑Augmented Generation)
Scene 1 — Early morning: Setting up SAS Retrieval Agent Manager
User need: A reliable foundation for answers.
Product help: Choose a Large Language Model (LLM), embedding model, and vector database.
Library analogy: Similar to setting a catalog system using embedding models and vector databases for storage the indexes.
Select any image to see a larger version.
Mobile users: To view the images, select the "Full" version at the bottom of the page.
Scene 2 — Mid‑morning: Ingesting sources
User need: Bring in new guidance without chaos.
Product help: Upload documents from files, retrieve them from Git or using a custom program. SAS Retrieval Agent Manager chunks and vectorizes content.
Library analogy: Like organizing books on shelves = collections. Vectorization goes through each book, stores the chunks in the vector database.
Scene 3 — Late morning: Evaluating accuracy
User need: Trustworthy answers before field use.
Product help: Run manual or automated evaluations producing a RAGAS (Retrieval‑Augmented Generation Assessment) score.
Library analogy: Reviewers = evaluation LLMs. Audits = RAGAS evaluations.
Scene 4 — Early afternoon: Chat with the library
User need: Fast, grounded answers for urgent questions.
Product help: Chat retrieves answers with citations and response metrics.
Library analogy: Librarians = chat LLMs.
Scene 5 — Afternoon: Assigning agents
User need: Turn updates into usable training materials, summaries, alerts, etc.
Product help: Agents generate summaries and structured outputs.
Library analogy: Specialized research assistants = agents.
Scene 7 — Late afternoon: Connecting systems
User need: Push updates into operational tools.
Product help: Integrate via MCP (Model Context Protocol) tools and APIs (Application Programming Interfaces).
Library analogy: Request forms = MCP tool templates. Back‑room service desk = MCP tool servers. Specialized functions = MCP tools. Checkout desk for external patrons = agents API access.
Scene 7 — End of the day: Automation
User need: Keep knowledge current without manual work when the source changes or a new vectorization job is triggered. Product help: Scheduled re‑indexing keeps collections fresh.
Library analogy: Scheduled re‑cataloging = automations.
Conclusion
SAS Retrieval Agent Manager turns static PDFs (and many other file types) into a living, trusted knowledge system. It retrieves the right sources and drives action.
The demo shows how SAS Retrieval Agent Manager pairs trusted content with retrieval, evaluation, and automation, at speed and scale.
Next steps
More posts will dive into each step, configurations, evaluation strategies, and code snippets.
Additional resources
Workshop
A Smarter Way to Unlock Unstructured Data: SAS Retrieval Agent Manager in Action
Subscriptions: SAS Decisioning Learning Subscription.
Posts in This Series
- Post 1 — The RAG Workflow: A Day in the Life of Ramona.
- Post 2 - Setting Up LLMs, Embeddings, and Vector Databases in SAS® Retrieval Agent Manager.
- Post 3 - Stocking the Library: Local Files, Custom Scripts, and Git Sources in SAS® Retrieval Agent Manager.
Posts on the same topic
- Unlocking agentic AI potential with MCP tools in SAS Retrieval Agent Manager
- Considerations for evaluating the effectiveness of RAG models
- Evaluating the Performance of Retrieval Augmented Generation Pipelines using the Ragas Framework
- How Retrieval Augmented Generation (RAG) Works?
Official SAS Documentation
Video Disclaimer
- The character name ‘Ramona’ is a playful nod to SAS® Retrieval Agent Manager (RAM). Any resemblance to real persons, is purely coincidental.
- The intro and conclusion scenes in the video were generated using SORA from Azure OpenAI.
- All other screenshots are real product screens from SAS® Retrieval Agent Manager stable version 2026.01. Real functionalities.
For further guidance, reach out for assistance.
Find more articles from SAS Global Enablement and Learning here.
Ready to see what SAS Viya Copilot can do?
Visit the Tips & Tricks page for setup guidance, demos, and practical examples that show how Copilot supports your workflows.
