Prompt Engineering is Dead; Build LLM Applications with DSPy Framework
OVERVIEW
EXPERIENCE | In Person |
---|---|
TYPE | Breakout |
TRACK | Generative AI |
INDUSTRY | Retail and CPG - Food |
TECHNOLOGIES | AI/Machine Learning, Delta Lake, Governance |
SKILL LEVEL | Intermediate |
DURATION | 40 min |
DOWNLOAD SESSION SLIDES |
Stop prompt engineering in LangChain. You wouldn’t hand-select weights of your neural network, so don’t hand-select your prompts. DSPy is an open-source framework that provides a paradigm shift towards building pipelines to optimize language model prompts, model tuning, and LLM applications with code. In this session executives will learn how adopting DSPy can save time and resources while enhancing application performance, and developers will leave equipped with knowledge on how they can incorporate DSPy into their LLM application development process. We will demonstrate how to move away from traditional prompt engineering to a more systematic approach - stitching together DSPy’s “signatures, modules, and optimizers” functionality to create a system that leverages language models for specific tasks within your application – with the goal of empirically optimizing your LLM application’s performance
SESSION SPEAKERS
Matt Yates
/Sr. Director AI, ML, & Data Science
Sephora