To compete in the age of GenAI, companies are applying language models to their own domain-specific data. They aim to boost productivity and enrich digital interactions while minimizing governance risks such as hallucinations. These domain-specific language models require new data pipelines that comprise text or other semi-structured data, vector databases, and custom applications.

This webinar offers a guide for data engineers, data scientists, and ML engineers that must design and manage these pipelines. It explores three primary approaches to supporting domain-specific language models: prompt enrichment, one-shot prompting, and retrieval-augmented generation (RAG). In each case data teams aim to provide language models with the facts and context they need to generate accurate, trustworthy, and governed responses that drive value for the business.

You will learn:
- What generative AI is and how it works
- Why companies are adopting domain-specific language models
- How data pipelines support these GenAI initiatives
- How prompt enrichment, one-shot prompting, and RAG compare with one another
- What criteria to use when evaluating these approaches

Sponsor: Matillion