Role requirements
Design, test, refine and document enterprise prompts for GPT, Claude and comparable LLMs. Demonstrate accuracy, safety, relevance, performance, RAG, agent workflows, and governance.
- Zero-shot, few-shot and role-based prompting
- CoT/ToT/ReAct concepts and verifiable outcomes
- Python, evaluation frameworks and agent orchestration
- RAG, vector search, citation quality and grounding
- Regression testing, safety, reproducibility and reporting
