Weaviate Podcast MIPRO and DSPy with Krista Opsahl-Ong! - Weaviate Podcast #103
11 snips
Aug 28, 2024 Krista Opsahl-Ong, a leading developer and scientist at Stanford University, is the mastermind behind MIPRO and DSPy. In this engaging discussion, she illuminates the world of Automated Prompt Engineering and its impact on language models. Krista dives into the challenges of manual prompt construction, highlighting innovative algorithms that streamline the process. The conversation also explores the intricacies of structured outputs, multi-stage language programs, and the revolutionary potential of Large Language Models in AI-driven solutions.
AI Snips
Chapters
Transcript
Episode notes
Manual Prompt Taught Krista Automation
- Krista worked with a clinician to extract insights from health records using manual prompt engineering.
- Manual prompt optimization was powerful but slow, inspiring her to develop automated prompt engineering.
Role of Automated Prompt Engineering
- Automated prompt engineering aims to find optimal prompts without manual trial and error.
- Optimizing prompts automatically addresses brittleness when the underlying model changes.
Power of Multi-Stage Language Programs
- Multi-stage language programs break complex tasks into modular language model or retriever calls.
- This decomposition enables more effective and controllable task solving than single LM calls.