Best Practices For Cost-Effective Development And Deployment On Snowflake ML

In this session from Summit 2025, two of Snowflake's most senior ML architects take a deep-dive into best practices for ensuring maximally efficient and cost-effective development and deployment of ML capabilities on Snowflake. They cover optimization techniques throughout the end-to-end ML workflow to achieve best-in-class price performance, as well as best practices for managing efficient model deployment and production pipelines. They also highlight different approaches and trade-offs as well as considerations for how to maximize performance and cost efficiency when running enterprise models on Snowflake ML.

CHAPTER BREAKS
00:00--Intro
00:41--Where Does ML Cost Come From in Snowflake?
01:43--Snowflake ML Simplifies Cost Governance
02:53--Integrated Set of Capabilities for end-to-end ML in Snowflake
03:42--Best Practices #1: Use Container Runtime
05:25--Best Practices #2: Scale Feature Development and Support Reusability
07:13--Best Practices #3: Scale Training Compute On-Demand
08:20--Best Practices #4: Load Data with DataConnector
09:05--Optimizing Production Pipeline
10:35: Best Practices #5: Deploy the Right Way
11:37--Best Practices #6: Infer Incrementally
13:26--Managing ML at Scale
16:32--Best Practices #7: Separate Dev & Prod
17:20--Best Practices #8: Budget & Monitor
17:59--Wrap Up

For more info on Snowflake ML, click here:
👉 https://www.snowflake.com/legal-files...

Subscribe to this channel for more great content!
👉 http://www.snowflake.com/YTsubscribe/

Click here to start your 30-day free Snowflake trial, which includes $400 worth of free usage:
👉 https://snowflake.com/youtube-dev-trial

Explore sample code, download tools, and connect with peers:
👉 https://developers.snowflake.com/

#Snowflake #AIDataCloud #AI #GenAI #ArtificialIntelligence