# Platform Governance & Performance Deep Dive

This course is built for engineers who already have working knowledge of Snowflake and want to move from getting queries to run correctly toward getting them to run efficiently, securely, and at a cost the business can live with. It moves past the basics of writing SQL against Snowflake and into the internals that separate a functional data platform from a well-run one: how compute and storage actually behave under load, how to read and act on a Query Profile, and how to put real governance and cost controls in place before they become a problem rather than after.

Each topic in this course pairs a foundational course from Snowflake University with original material built specifically for this program. The Snowflake courses give you the grounding Snowflake itself considers essential. The original lessons go further, covering the parts of day to day platform work, disk spilling, exploding joins, clustering depth, resource monitor configuration, and Snowflake Horizon governance features, that sit just past where most introductory material stops.

## What You'll Learn

The course opens with the architectural foundations that everything else depends on: how Snowflake separates storage from compute, how data is physically organized into micro-partitions, and how that organization determines whether a query prunes efficiently or scans far more than it needs to. From there it moves into the practical mechanics of running a platform day to day, structuring databases and schemas so a project stays navigable as it grows, setting up roles and access control the way Snowflake's own security model expects, and applying governance features like object tagging, dynamic data masking, and row level security to protect sensitive data without maintaining a maze of separate views. The course also introduces Snowpark and Python for data engineering work, and building lightweight data applications with Streamlit directly on top of Snowflake data.

The second half of the course turns to performance and cost. You'll learn to read a Query Profile well enough to diagnose why a specific query is slow, distinguishing a genuine compute problem from a data organization problem, and to recognize and fix the patterns that cause the most pain in practice, including local versus remote disk spilling, exploding joins, and metadata pruning failures. On the cost side, you'll work with Resource Monitors and the FinOps framework of visibility, control, and optimization, and cover the operational safety net of Time Travel, Fail-safe, and database cloning, along with the collaboration and data sharing capabilities Snowflake provides for working across teams and organizations.

## Who This Course Is For

This course assumes you can already write SQL and have used Snowflake (Or similar data platform) in some capacity, whether that's building pipelines, writing analytics queries, or administering an account. It does not assume prior exposure to performance tuning, cost governance, or Snowflake's security features specifically. If you are completely new to Snowflake, the first unit of each subsection points to the appropriate beginner course to complete before continuing.

## How the Course Is Structured

The course is organized into two main sections.

The first covers foundations, administration, and Python:

• Architecture and compute mechanics
• Container hierarchy and project structure
• A hands-on data warehousing workshop
• Accounts and governance
• Snowpark and Python
• Building data applications with Streamlit.

The second covers query optimization and FinOps:

• Query history and caching
• Query profiling and bottleneck remediation
• Resource monitors, cost management
• Data protection and recovery
• Sharing and marketplace collaboration.

Each subsection opens with the relevant Snowflake University course as a starting point, followed by original lessons that go deeper into the parts most relevant to running a platform well.

## Time Commitment and Completion

Plan for roughly 30 hours per section. Completing the course involves finishing the linked Snowflake courses and original lessons, submitting hands-on labs executed in the Factored Snowflake account, an interactive AI interviewer session, and a final Thebai assessment.