Join us for the final installment of our AI Tech Talk series with Luis, where we bring together everything we've learned so far to explore one of today’s most powerful AI techniques: Retrieval-Augmented Generation (RAG).
In Part 1, we uncovered the foundations of Transformer Models and Distillation, exploring how neural networks evolved into the large language models (LLMs) we use today. In Part 2, we went hands-on with Fine-Tuning and Distillation, learning how to create and deploy custom LLMs tailored to specific use cases.
Now, in Part 3, we’ll see how to make these models even smarter, more accurate, and grounded in real-world data. Luis will introduce RAG, explaining how it enhances LLMs by reducing hallucinations and keeping responses current through verifiable information retrieval.
You’ll also gain practical insight into how RAG works under the hood — including vectorization, distance measurements, and context grounding — and see a real-world use case demonstrating how RAG can improve community well-being by automating municipal workflows.
Agenda
- What is RAG?
- Why RAG?
- Grounding and Context
- Vectorization
- Distance Measurements
- Real-world Use Case
Whether you’ve joined the earlier sessions or you’re attending for the first time, this talk will equip you with a solid understanding of how RAG can make AI systems more reliable, transparent, and actionable in practice.
