LLM optimization, evaluation frameworks, COSTAR/chain-of-thought. Craft and optimize prompts for maximum AI performance.
- Deep understanding of LLM behaviors, quirks, and prompt templates
- Experience with systematic testing and evaluation of prompts (e.g., PromptFlow)
- Familiarity with advanced prompting techniques (CoT, ReAct, Few-Shot)
- Ability to collaborate with engineers to integrate prompts into codebases