Workshop on LLM Techniques: RAG, Guardrails and MoE [Technical]
GPTechday is a leading global AI community, bringing together AI engineers, researchers, founders, investors, and enthusiasts so we can stay ahead of the AI wave.
Bring your laptop as we will be coding and building real RAG workflows, guardrails, and MoE architectures!
Workshop Instructor:
Muntaser Syed,
GPU Developer Advocate, Ex-NVIDIA
PhD Candidate, Florida Institute of Technology
Details:
Title: "How to Use LLM Techniques: RAG, Guardrails and MoE"
Description:
💡 Join us for a fast‑paced workshop that demystifies three of today’s most practical large‑language‑model techniques—Retrieval‑Augmented Generation (RAG), Guardrails, and Mixture‑of‑Experts (MoE). Through bite‑sized lectures and live coding demos, you’ll learn how to fuse vector search with LLMs to ground answers in your own data, wrap models in robust policy and schema constraints to tame hallucinations, and tap sparse‑routing MoE architectures to boost performance without breaking the compute budget.
Whether you’re a developer looking to productionize chatbots, an architect evaluating scaling strategies, or simply curious about the nuts and bolts behind cutting‑edge AI, you’ll leave with reusable notebooks, a starter repo, and a clear mental model of when—and when not—to wield these tools.
Agenda:
1:00 - 1:30 PM: Check-in and Getting Ready
1:30 - 4:00 PM: Workshop Training
4:00 - 4:30 PM: Extended Support & Q&A
4:30 - 5:00 PM: Networking & Closing