senior-prompt-engineer

Design robust LLM prompts for production AI by enforcing structured templates, establishing evaluation criteria, and applying advanced patterns like Chain-of-Thought, ReAct, and Tree of Thoughts to elicit consistent behavior.

The best way to structure RAG integration prompts is to design retrieval-augmented generation pipelines that guide the LLM to invoke retrieval tools for fresh data while summarizing documents with key decisions and actions.

Build and evaluate agent architectures by designing tool interfaces for autonomous workflows and establishing strict metrics and evaluation procedures to test prompts and responses across enterprise contexts.

Use advanced prompting patterns like ReAct and Tree of Thoughts when defining prompt-driven LLM application architectures that require complex orchestration and autonomous workflows in enterprise AI products.

Test and establish evaluation metrics for LLM responses by creating structured procedures that measure prompt reliability, response accuracy, and tool invocation success across various enterprise AI contexts.

Yes, you can design prompt-driven workflows for SaaS assistants by utilizing RAG system design to surface up-to-date information and integrating tool interfaces to handle autonomous tasks.