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

Get Started with the Snowflake Connector for Google Analytics Reference Architecture: https://lnkd.in/geUCGsYs This solution architecture helps you build and deploy a BI dashboard visualizing user analytics from websites. We use the Snowflake Connector for Google Analytics to synchronize data from GCP into Snowflake. We view the raw data using Snowflake Notebooks and visualize the aggregated data using Streamlit.

Deep dive on Arctic Agentic RAG: https://lnkd.in/g8gNfmd8 Highlights: 1️⃣ VerDICT: A Smarter Approach to Query Clarification Instead of retrieving documents for every possible interpretation, VerDICT integrates verification upfront, ensuring only grounded and answerable interpretations are used—reducing inefficiency while improving accuracy. 2️⃣ Enterprise-Grade Applications Whether it's customer support automation, compliance, internal knowledge management, or analytics, Arctic Agentic RAG enables precise retrieval by dynamically clarifying vague queries before generating responses. 3️⃣ Open-Source Innovation We're making Arctic Agentic RAG accessible to researchers and practitioners, providing lightweight, modular tools to experiment, prototype, and push RAG advancements further.

We have plenty of new features for Cortex Search that offer a scalable and customizable foundation for search and agentic applications built on Snowflake data. 1️⃣ Increased scale and affordability 2️⃣ Improved customizability 3️⃣ New preview features, like Cortex Search Admin UI See what’s new: https://lnkd.in/gkfyrV9W

Cortex Analyst is now generally available! What’s new: 1️⃣ Handling increased schema complexity 2️⃣ Semantic model generation and monitoring 3️⃣ Customization for business-specific logic 4️⃣ Proven performance on benchmarks Full updates: https://lnkd.in/gkfyrV9W

Snowflake Cortex Search delivers exceptional search quality out of the box, outperforming current enterprise search tools on a broad set of benchmarks covering scenarios such as product, email, technical and web search by up to 15% better NDCG@10. Learn how Cortex Search empowers you to achieve superior results with less time, less effort and greater confidence: https://lnkd.in/gZvXPyA2

Get started building data engineering pipelines using Snowpark in Snowflake Notebooks. In this session, we’ll show you how to build a: - Data share from the Snowflake Marketplace to access weather data - Data engineering pipeline with a Notebook to ingest Excel files into Snowflake - Data engineering pipeline with a Notebook to transform and aggregate data - DAG (or Directed Acyclic Graph) of Tasks to orchestrate/schedule the pipelines - CI/CD pipeline to deploy the Notebooks to production Join us February 12: https://lnkd.in/gZvF7TMQ

Build an intelligent Q&A system using Anthropic's Claude and Snowflake Cortex. In this demo, you'll create an end-to-end application that can process PDFs, search through documents, and answer questions using AI. Here are the steps: 1️⃣ Setting up your Snowflake environment 2️⃣ Processing PDFs with Cortex 3️⃣ Creating a vector search system 4️⃣ Building a chat interface with Streamlit 5️⃣ Integrating Claude for intelligent responses Watch now: https://lnkd.in/gu_keqit

With Snowflake Cortex, you can streamline support case analysis. Our latest solution integrates Cortex LLMs, Cortex Search, and LangChain to deliver actionable insights, faster resolutions, and an intuitive chatbot experience. Key features include: ⚡️ Lightning-fast ticket summarization 💡 AI-driven insights 🧠 Enhanced intelligence with LangChain Get started on the solution: https://lnkd.in/g4W2SPjQ Quickstart: https://lnkd.in/gyvq4km8

Benchmarking LLM-as-a-Judge for the RAG Triad Metrics. In this blog, we share the results of benchmarking the three LLM Judges on standard ground truth data sets — TREC-DL4 for context relevance, LLM-AggreFact for groundedness and HotpotQA4 for answer relevance. ✅ The TruLens Groundedness judge tops the field on the F1 metric, balancing both precision and recall ✅ Our Context Relevance judge also takes the top F1 score, beating out UMBRELA and an adjacent open source evaluator ✅ The LLM Judge for answer relevance is comparable to the MLflow Judge Learn more: https://lnkd.in/dgbwM-iJ