Prompt Optimizer with Custom Metrics
Independent PiSkill directory guide. The original prompt remains hosted by Google Cloud Generative AI.
What does this prompt do?
Shows how to optimize a Gemini prompt against a task-specific evaluation metric instead of relying on a generic measure of answer quality.
Primary use case
Improve a prompt using a custom metric aligned with the application's requirements.
Expected output
A prompt variant optimized against the supplied quality criterion.
Inputs or variables
- prompt
- examples
- custom metric
Related prompts
Vertex AI Prompt Optimizer
Demonstrates how to improve prompts with the Vertex AI Prompt Optimizer and evaluate whether optimized variants perform better for a target task.
Gemini Prompt Design Introduction
Provides a practical notebook for learning and applying prompt-design techniques with Gemini, including clear instructions, examples, context, and output constraints.
Long Prompt Optimization
Applies Vertex AI prompt optimization to long instructions where structure, retained constraints, examples, and context use need careful evaluation.
AI-Assisted Data Science
Uses Gemini to explore a data question, inspect a dataset, develop analysis steps, produce code, and explain findings with assumptions and validation checks.
BigQuery AI Operators
Demonstrates BigQuery AI operators for applying model capabilities to warehouse data with explicit SQL, input columns, result handling, and cost-aware validation.
BigQuery RAG with Gemini
Builds a retrieval-augmented question-answering workflow with BigQuery data, embeddings, retrieval, prompt context, and grounded Gemini responses.