Snowflake Developers

Snowflake Cortex Agents, now in public preview! Cortex Agents orchestrates across structured and unstructured data for accurate AI-driven decisions from within the secure Snowflake perimeter; Cortex Agents use Cortex Analyst (now in GA) and Cortex Search as tools. Swipe 👈 to see what’s new. Anthropic’s Claude 3.5 Sonnet is used by Cortex Agents to deliver accurate, efficient and governed data insights at scale. All the details: https://lnkd.in/gkfyrV9W

Excited to share our work on Speculative Decoding @Snowflake AI Research!  🚀 4x faster LLM inference for coding agents, 🚀 2.4x faster LLM inference for conversational use cases, ✅ Available via Arctic Inference as an easy-to-install plugin for vLLM!   Arctic Inference unifies several innovations into a powerful hybrid speculator: - Suffix Decoding for fast speculation of long, repetitive sequences (20μs per token on CPU) - Lightweight draft model pipeline using MLP/LSTM with seamless training-to-deployment - System-level vLLM optimizations that minimize speculation overhead 🔍 Learn more in our blog: https://lnkd.in/gR9GMkKE 📦 Everything is open-sourced: - Arctic Inference: https://lnkd.in/gEWenC-J - Arctic Training: https://lnkd.in/gRP-fu4Y Arctic Inference is implemented as a vLLM plugin, so it’s super easy to get started!   🙌 Huge thanks to Ye W., Gabriele Oliaro, Jaeseong Lee, Aurick Qiao, Samyam Rajbhandari and many others for 6+ months of incredible work!

Level up your Python data pipeline skills with Snowflake! 🚀 This free, 1-hour webinar offers hands-on experience building end-to-end, scalable data workflows. We'll dive into: ➡️ Ingesting data (S3 to Snowflake) ➡️ Processing with Snowpark DataFrames (joins, aggs) ➡️ Visualizing insights with Snowflake Notebooks ➡️ Using pandas on Snowflake (even reading from Git!) ➡️ Automating pipeline execution Save your spot: https://lnkd.in/gw3uaAFp

📢 #SnowflakeDevDay is back! Join us June 5 and get ready for: 🎤 Luminary talks with LandingAI's Andrew Ng, Anthropic's Jared Kaplan, and Lisa Cohen, and more 💻 Hands-on labs 🚀 A first look at the latest developer tools and innovations 🤝 Opportunities to network with peers and Snowflake experts Register for free today: https://lnkd.in/gW9ehw2a

Apache Iceberg™ 1.9 is here! ❄️ Join our Lead Developer Advocate for OSS Danica Fine for a quick walkthrough of what’s new: - Core changes and new features - Spec updates including v3 progress - Deprecations and removals Learn more: https://lnkd.in/gj3iKxsk Release notes: https://lnkd.in/gr9Tdr6Y

Learn how Moser Consulting reimagined their data platform, Honeycomb, by consolidating onto Snowflake, achieving a 70% reduction in data platform costs while boosting performance, scalability, and governance. Honeycomb’s original architecture was robust, but also complex. Moser saw the opportunity to simplify and scale by going all-in on Snowflake, leveraging features such as Snowpipe, Snowpark, and Cortex AI. By transitioning to a Snowflake-centric architecture, Moser: - Eliminated platform redundancies and overhead - Improved developer efficiency - Unified data governance and security under one platform - Realized major cost savings with usage-based pricing This is a great example of how organizations can future-proof their data strategy and drive innovation with Snowflake’s single, integrated platform. Dive into their full journey: https://lnkd.in/eHm6V475

Leverage GPUs in Snowflake Notebooks to supercharge your machine learning workflows in under 5 minutes. In this video, Senior Developer Advocate James Cha-Earley shares: ✅ Why GPUs accelerate machine learning workflows ✅ How to select GPU-enabled container images ✅ Tips for choosing the right GPU size for your workload ✅ Best practices for monitoring GPU resource usage ✅ Cost management strategies for GPU sessions Learn more about Snowflake's machine learning offerings 👉 https://lnkd.in/gSZzsEei

Smarter Defect Detection in Snowflake ML with Computer Vision. We’re making AI-powered Quality Control easier than ever! Using Snowflake Notebooks on Container Runtime, we’ll train a multiclass defect detection model—boosting manufacturing efficiency while reducing costs and resource waste. 🔍 How it works: ✅ Preprocess annotated images ✅ Convert images to base64 and store in Snowflake tables ✅ Train a PyTorch model on multi-GPUs ✅ Register the trained model in Snowflake Model Registry ✅ Deploy a Streamlit app to run detections & visualize results Watch the demo: https://lnkd.in/g9zmKg3A Try for yourself here: https://lnkd.in/gpuuKN_Q

Managing access in Snowflake just got easier. Learn how Role-Based Access Control (RBAC) with database roles can help you simplify permissions and strengthen your security posture. Check out #DataSuperhero Maja Ferle's blog: https://lnkd.in/gb2fQ2xV

Get started with the SnowConvert Migration Assistant and accelerate your post-conversion cleanup in your SQL code. Here's how to begin: - Install Snowflake VS Code extension - Sign in to Snowflake with VS Code extension - Enable SnowConvert Migration Assistant - Open a folder containing SnowConvert migration results - See issues and click the lightbulb to help resolve - Refine results Full documentation: https://lnkd.in/eg9Fjaeu