Matt Hammond’s Post

You know the saying: “Money doesn’t change you. It just makes you more of what you already are.” Same goes for AI. If you don’t have a data strategy, AI won’t magically make it all better for your organization. Snowflake gets that. That’s why we built our platform to eliminate silos, create unified governance and security and optimize total cost (TCO) across workloads. Get that right, and you’ll be ready for Snowflake’s ready-to-use and easy-to-use AI and ML: - LLMs for summarize, classify and other NLP tasks - ML functions for forecasting, anomaly detection, contribution exploring and classification - End-to-end MLOps for traditional data science models - Snowpark container services for full-stack GenAI apps on fully-managed GPUs How do you strengthen your data strategy and accelerate success with AI? Governance is a great place to start. Learn more about Snowflake Horizon in this outstanding webinar series starting July 9:

“Snowflake is expensive.” How? Snowflake compute is the most efficient in the industry. (That’s what put us on the map for data warehousing). That same engine also runs data engineering and AI workloads — in SQL, Python, Java and Scala. We outperform Spark and managed Spark across workloads. (It’s not close). - 4.6x faster runtimes - 34% reduced cost - 50% less ops burden Snowflake’s fully-managed service also handles provisioning, scaling and suspending our compute. Meaning: less wasteful spend. Same goes for GPUs, for compute-intensive workloads. (See: Snowpark container services). We could do this all day. Snowflake even saves customers money just from ongoing performance enhancements. We measure how much faster and cheaper we’re running, with the Snowflake Performance Index. How much faster? 27% since August 2022. Meaning: Snowflake customers get an additional 13% in savings every year on average. If you’re hearing that Snowflake is expensive, it might make sense to consider the source. If you talk to Snowflake customers, they’ll tell you a different story. Just ask IGS Energy, who saved 75% for AI and ML by leaving Databricks for Snowflake.

Chatted with a CIO who thought his organization was “about a year away from being ready for AI.” Doesn’t have to be that way. Snowflake makes it easy for you to get to the cloud, create a single governance and security layer and activate your teams to activate your data. Snowflake Cortex AI, for ready-to-use LLMs and ML functions for citizen data science. Snowpark ML, for MLOps without the ops burden. Snowpark container services, for compute-intensive GenAI with fully-managed GPUs. What you don’t need to waste your energy on with Snowflake: infrastructure, platform, compute and storage maintenance. “It just works.” Check out this developer guide on how to create enterprise Call Center Analytics in Snowflake using OpenAI Whisper, Snowflake Cortex LLMs, Snowpark container services and Streamlit. Takes only 35 minutes.

“It just works”

Snowflake’s Head of AI/ML for Financial Services Jonathan Regenstein is hosting an outstanding event on building an enterprise AI strategy today at 12 pm CT. ✅ Design considerations unique to financial services ✅ Today’s LLM landscape, including Snowflake Arctic ✅ Impact GenAI use cases for financial services ✅ How you should expect the AI landscape to evolve Also featuring a live demo of Arctic, the LLM for enterprise intelligence, breakthrough efficiency and true openness — built by Snowflake AI. Register here:

42% of CIOs plan to have GenAI in prod by Q3. Is that realistic? Not if you don’t clear needless blockers. Snowflake eliminates the cost, complexity and risk that gets in the way of your GenAI strategy. No bloated LLMs that drive AI overspend. No fighting with complicated GPU infrastructure. No sending data outside your governance and security framework. Only the industry’s most targeted data and AI platform, so you can focus on what actually matters: business outcomes and enterprise intelligence. Learn more about how to deliver AI in seconds, apps in minutes and fully custom in hours with Snowflake Cortex: https://lnkd.in/g2qvYPqQ