Description: Move beyond basic prompting and learn how to apply AI effectively to real-world HPC workflows. This five-part webinar series explores how large language models use context, how to ground AI responses in reliable information, how to integrate AI with codebases and operational systems, and how to transform AI-generated output into secure, reviewable work. Participants will also discover how agentic AI frameworks can automate complex tasks while operating entirely within authorized HPCMP environments.

Topics Covered:

• Applied AI I: How LLMs Use Context & Tool Calling – Learn how language models process context, why they fail, and how structured prompts produce more reliable results.

• Applied AI II: RAG, Fine-Tuning, and Grounding – Explore retrieval-augmented generation (RAG), embeddings, semantic search, and techniques for producing accurate, source-grounded AI responses.

• Applied AI III: AI for Codebases, APIs, and HPC Workflows – Discover how AI interacts with code repositories, APIs, operational data, and tool connectors to solve real engineering problems.
• Applied AI IV: From AI Output to Reviewable Work – Develop a repeatable workflow for evaluating, refining, and validating AI-generated content using human review, security best practices, and governance controls.
• Applied AI V: Agentic AI Frameworks in Secure Boundaries – Learn how autonomous AI agents coordinate work across HPC resources while remaining within secure, authorized computing environments.

Controlled by: DoD HPCMP Controlled by: PET Program CUI Category: OPSEC Limited Dissemination Control: FEDCON POC: Mr. Ronald Hedgepeth, pet@hpc.mil

---

CUI