Skill UI

This comprehensive skill bridges the gap between initial idea discovery and final model deployment. It ingests a structured experiment plan (EXPERIMENT_PLAN.md), automatically implements the required code, performs cross-model code review, deploys the setup to GPU, and collects initial quantitative results, allowing for a rapid iteration loop in AI research.

This guide identifies and provides solutions for common anti-patterns and mistakes when working with Snowflake. It covers best practices in SQL writing, warehouse resource management, data loading strategies (e.g., using MERGE), and access control setup. It is essential for auditing existing deployments, onboarding new users, or performing code reviews to ensure cost efficiency and data integrity.