Gyumin Choi

Snowflake — Enterprise AI Deployment

Challenge: Tier-1 Korean conglomerates — retail, financial, and construction — wanted to modernize onto an AI-ready data platform, but every deal came down to whether the technical value could be proven quickly and then actually adopted by their engineering teams.

Solution: Acted as the primary technical authority across the full customer lifecycle for accounts including GS Retail, GS E&C, Musinsa Payments, and Hyundai Department Store. Designed and executed hands-on POCs, engineered an end-to-end MLflow-to-Snowflake production migration framework inside a one-week window, and built an enterprise agentic AI framework on Model Context Protocol (MCP) with custom Python orchestrators — deploying autonomous coding agents that refactored 100+ complex dbt models in 6 hours. Followed through with enablement: the featured AI/ML keynote at Snowflake World Tour Seoul and deep-dive notebook and pipeline workshops for senior engineering teams.

Impact: Secured a $250K enterprise contract head-to-head against Databricks, validated 5-10x data processing efficiency gains against legacy cloud platforms in POC, and delivered the event's #1 voted session to 400+ technology executives.

Stack

SnowflakeData ArchitecturePythonPrompt EngineeringLLM EvaluationAWS

Deployments

Screenshots

© 2026 Gyumin Choi