Sigma Skills Coming to Snowflake Cortex Code

By popular demand, Sigma Skills are coming to Snowflake Cortex Code ❄️ This is a great step forward for teams already building in Snowflake. Data engineers will be able to create, update, and validate Sigma assets directly from Cortex Code. In practice, that means you can: 1. Create Sigma data models on top of Snowflake tables 2. Define metrics, relationships, and business logic 3. Update models when schemas change 4. Build governed workbooks and dashboards 5. Automate report delivery and downstream workflows ⚙️ What I like about this is that it meets data teams where they already work. Snowflake remains the foundation. Cortex Code becomes the place to build and manage the workflow. Sigma brings the governed analytics, reporting, AI app, and business-user experience on top 📊 For Snowflake customers thinking about Cortex, AI-powered data workflows, or governed self-service analytics, this is well worth a look. #Snowflake #CortexCode #SigmaComputing #Analytics #AI #DataApps

🚀 Snowflake is evolving FAST — and 2026 is all about AI + Data + Simplicity Here are some of the most exciting recent updates that are reshaping the Snowflake ecosystem 👇 🔹 AI is now deeply integrated With Snowflake Cortex, organizations can build and run LLM-powered applications directly within Snowflake — no data movement, fully governed 🔐 🔹 Cortex AI Functions & AISQL Run AI directly inside SQL queries — blending structured + unstructured data seamlessly 🤯 🔹 Cortex Code (AI-powered development) Developers can now write, optimize, and debug queries using AI assistance inside Snowsight ⚡ 🔹 Dynamic Tables & Iceberg enhancements Better control over data pipelines with improved refresh, partitioning, and open table format support 📊 🔹 Stronger Data Governance & Security New features like data classification, access history improvements, and Trust Center updates ensure enterprise-grade security 🔒 🔹 Snowpipe Streaming Error Logging Better observability = faster debugging and reliable ingestion pipelines 🚀 🔹 New Data Types for AI workloads Enhanced VECTOR type, structured data types, and FILE support enable modern AI/ML use cases 📦 --- 💡 Key Takeaway: Snowflake is no longer just a data warehouse — it’s becoming a complete AI Data Cloud platform. From data engineering → analytics → AI → apps Everything is now happening in ONE place. --- 🔥 If you're working in Data Engineering / Snowflake: 👉 Time to upskill in Cortex AI + Snowpark + AI SQL #Snowflake #DataEngineering #AI #GenerativeAI #DataCloud #SnowflakeCortex #BigData #Analytics

The future of data engineering is not just about building pipelines, it’s about making them smarter, faster, and more adaptable. This is a great example of how GenAI is starting to reshape the modern data stack. Translating dbt models across SQL dialects has always been a friction point, especially in multi-cloud or hybrid environments. With tools like dbt-sqlx powered by GenAI, that complexity can be significantly reduced. What stands out here is the shift from manual transformation logic to intelligent automation. Instead of rewriting queries for different platforms, engineers can focus more on data modeling, governance, and delivering business value. This is where data engineering is heading: Less time on repetitive SQL rewrites More focus on scalable data design Faster cross-platform adaptability Stronger alignment between engineering and analytics As GenAI continues to integrate into tools like dbt, Snowflake, and Databricks, the role of a data engineer is evolving from pipeline builder to platform enabler. Curious to see how teams start adopting this in real production workflows and what it means for standardization across data platforms. #DataEngineering #dbt #GenAI #ModernDataStack #DataTransformation #AnalyticsEngineering #Snowflake #Databricks #BigQuery #ELT #DataOps #AIinData #DataInnovation

Still manually converting SQL during client calls? There’s a better way. Excited to share a solution we recently built — a SQL Conversion Accelerator powered by Snowflake Cortex AI & Cortex Code 🚀 In real-world consulting, we often face a gap: 👉 Quick, ad-hoc SQL conversion needs 👉 vs heavy enterprise migration tools This accelerator bridges that gap by enabling instant, intelligent SQL conversion with: ✔ Side-by-side diff view ✔ Clear conversion explanations ✔ Zero setup — built for real consulting scenarios 💡 Built using Snowflake Cortex + Cortex Code 💡 Designed for workshops, RFPs, troubleshooting & learning 💡 Complements SnowConvert AI for enterprise-scale migrations Acknowledging the guidance and support from: 👏 Pallavi Sharma— for mentoring and guiding Cortex Code learning 👏 Soham Banerjee Krishna Mohan Bommi — for coordinating learning batches and creating a structured plan for cortex code enablement 👏 Sweety Mathew — for leadership and continuous support Would love to hear your thoughts and feedback! 📖 Full blog (link in comments 👇) #Snowflake #SnowflakeCortex #DataEngineering #AIinData #GenerativeAI #AIForDevelopers #SQL #CloudData #DataArchitecture #CortexCode

Enrolled and Attended Snowflake's Data for Breakfast and this one is worth your attention. Snowflake is conducting "Data for Breakfast: Making AI Real for Business" a virtual event built for every professional in the data ecosystem. The agenda is sharp and highly relevant: 🔹 Deep dive into Cortex Code: Snowflake's new AI agent transforming how Data Engineers, Data Scientists, Analysts, Software Developers and even Business Users interact with data 🔹 Hands-on lab sessions: practical, workflow-driven learning designed to build real working knowledge, not just theoretical exposure 🔹 Snowflake community connect: an opportunity to hear and connect with some of the brightest minds like Louis Lee, Sho Tanaka, Bharath Suresh, Majid Miri, Wisarut Sirimart, Anudeep Ayinaparthi and the snowflake community As someone building at the intersection of AI automation, Big Data and enterprise analytics, staying ahead of platform-level AI innovation is not optional. It is strategic. Cortex Code represents exactly the kind of agentic AI capability that will redefine how enterprises consume and act on data intelligence. If you are serious about AI in the data space this is a conversation worth joining. #Snowflake #DataforBreakfast #CortexCode #AIDataCloud #DataAnalysis

🚨 𝗠𝗼𝗱𝗲𝗹 𝗗𝗲𝗽𝗿𝗲𝗰𝗮𝘁𝗶𝗼𝗻𝘀 𝗔𝗿𝗲 𝗖𝗼𝗺𝗶𝗻𝗴. 𝗔𝗿𝗲 𝗬𝗼𝘂 𝗥𝗲𝗮𝗱𝘆? If you're building GenAI applications inside Snowflake Cortex or external functions, this is not just an announcement — it’s an architectural checkpoint. The following models are scheduled for deprecation: ⚠ GPT-4-turbo — April 16, 2026 ⚠ Claude Sonnet 3.5 — April 28, 2026 ⚠ Snowflake Arctic — April 28, 2026 🔎 𝗪𝗵𝗮𝘁 𝘁𝗵𝗶𝘀 𝗿𝗲𝗮𝗹𝗹𝘆 𝗺𝗲𝗮𝗻𝘀 𝗳𝗼𝗿 𝗗𝗮𝘁𝗮 & 𝗔𝗜 𝘁𝗲𝗮𝗺𝘀: • Review model dependencies in Cortex workloads • Validate prompt compatibility with newer model versions • Benchmark performance, latency & cost before migration • Update governance + approval workflows • Communicate impact to product stakeholders early 𝗠𝗼𝗱𝗲𝗹 𝘂𝗽𝗴𝗿𝗮𝗱𝗲𝘀 𝗮𝗿𝗲 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 “𝘃𝗲𝗿𝘀𝗶𝗼𝗻 𝗯𝘂𝗺𝗽𝘀.” They impact accuracy, token usage, pricing, and output behavior. 𝗛𝗶𝗴𝗵-𝗺𝗮𝘁𝘂𝗿𝗶𝘁𝘆 𝘁𝗲𝗮𝗺𝘀 𝘁𝗿𝗲𝗮𝘁 𝗟𝗟𝗠 𝗹𝗶𝗳𝗲𝗰𝘆𝗰𝗹𝗲 𝗺𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 𝗹𝗶𝗸𝗲 𝗶𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝗹𝗶𝗳𝗲𝗰𝘆𝗰𝗹𝗲 𝗺𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁. If your AI stack is inside Snowflake, now is the time to: ✔ Audit ✔ Test ✔ Upgrade ✔ Optimize 𝗗𝗼𝗻’𝘁 𝘄𝗮𝗶𝘁 𝗳𝗼𝗿 𝗯𝗿𝗲𝗮𝗸𝗶𝗻𝗴 𝗰𝗵𝗮𝗻𝗴𝗲𝘀 𝘁𝗼 𝗵𝗶𝘁 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻. 💬 Are you proactively managing model lifecycle — or reacting when Finance or Engineering flags issues? #Snowflake #GenAI #LLMOps #DataEngineering #AIArchitecture #ModernDataStack #Cortex #ArtificialIntelligence #CloudComputing #TechLeadership #DataOps

Spent some time last week going through Snowflake’s Data for Breakfast sessions and Northstar workshops ❄️ Covered areas across the stack: • Ingestion, transformation, and delivery • Declarative pipelines with Dynamic Tables • Building data applications and integrating AI workflows • Extending into end-to-end agent-style pipelines Good to see how Snowflake is pushing AI closer to the data layer, especially with Cortex and semantic layers. For systems handling sensitive data, being able to build AI agents directly on top of secure, governed enterprise data without moving it around is a game changer. A lot of this maps to problems I’ve worked on around ETL pipelines and RAG systems, just with a different approach to orchestration and architecture. Planning to test this further in a production-style setup. #Snowflake #DataEngineering #CortexAI #AIEngineering #MLOps #DataArchitecture #DynamicTables #AgenticAI #GenAI

I built tools that would take a data engineer 2-3 weeks in one afternoon using Snowflake's Cortex Code ❄️. 3 prompts. Plain English. Ready to share across my team. Here's what came out the other side: → A query optimization hub that scores your SQL, explains every performance issue, and rewrites the query in one click → A natural language chatbot your analysts can use to answer their own questions: no tickets, no waiting → A data quality command center that catches bad data before your VP does The best part? Every governance policy, RBAC, and data masking rule you've already built in Snowflake carries through every tool and agent, automatically. These tools are now just a prompt away. This is what I think the future of data engineering looks like. 👇 Full demo in the video below. #Snowflake #DataEngineering #CortexCode #AI #DataQuality #CoCo

If a photo is worth a thousand words, a video demo must be worth a million. In this case though I don't understand the underlying technology (that's ok I'm not the target audience) this video demo drive home the benefits of this new tool in a way that I wouldn't have understood through text alone. Seeing is believing. Thanks Jackson Cavenaugh for bringing it to life.

Exciting times in the Data Engineering space! 🚀 I’m thrilled to share a recent architectural win at NuWare where we’ve completely reimagined #CustomerDeduplication using the power of Snowflake’s AI-native features. Managing data at scale often means dealing with the "noisy" reality of duplicate records. Traditional fuzzy matching can be expensive and rigid, but by leveraging Snowflake Cortex, we’ve built a high-precision, low-latency deduplication engine that is both cost-effective and highly scalable. 🛠️ The Architecture: * Address Standardization: We first clean the "noise" by standardizing global addresses to ensure a consistent baseline. * Snowflake Cortex Batch Search: Using Cortex’s specialized search services to quickly identify potential match candidates across millions of rows. * The Hybrid Engine: * Cosine Similarity: For high-speed mathematical comparison of vector embeddings. * Low-Cost LLMs: We route complex cases to smaller, efficient LLMs within Snowflake to handle nuanced logic without the "Enterprise" price tag. * Smart Bypasses: Designed logic to automatically "bypass" clear non-matches, focusing compute resources only where they add value. * Automation: The entire pipeline is orchestrated seamlessly using Snowflake Tasks and Services, ensuring our "No Match" and "Golden Record" logic runs autonomously. 📈 The Result: A robust, "Agentic" approach to data quality that significantly reduces manual intervention while keeping compute costs optimized. Proud of how we are pushing the boundaries of the Modern Data Stack to solve age-old data problems with next-gen AI. Naveen Narayana Subramanian Kris #DataEngineering #Snowflake #Cortex #LLM #DataQuality #NuwareSystems #AI #MachineLearning #CloudData #BigData