How a financial services leader adopted Snowflake with CloudHive

🚨 Snowflake organization users: Uncovering Hidden Pitfalls🚨 In my latest Medium blog, I dive deeper into Snowflake Organization Usage feature - beyond the documentation and into real-world behavior observed during advanced testing. 🔍 Key findings: Stale role grants persist even after org group removal. Duplicate role entries appear after re-importing org groups. Insufficient telemetry in the Organization Account makes troubleshooting harder than it should be. These edge cases raise important questions about governance, automation, and audit reliability in multi-account Snowflake setups. 📖 Read the full breakdown here: https://lnkd.in/dQfmezQz 💬 I’d love to hear your thoughts. Have you encountered similar quirks? #Snowflake #DataGovernance #CloudDataPlatform #RoleManagement #Infosys #MediumBlog #MySnowflakeInsights

❄️ Snowflake Project - 1 Excited to share my latest Snowflake project: 𝗔 𝗿𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗱𝗮𝘁𝗮 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲 𝗳𝗲𝗮𝘁𝘂𝗿𝗶𝗻𝗴 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗧𝗮𝗯𝗹𝗲𝘀 𝗮𝗻𝗱 𝗽𝗿𝗼𝗮𝗰𝘁𝗶𝘃𝗲 𝗲𝗿𝗿𝗼𝗿 𝗵𝗮𝗻𝗱𝗹𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗦𝗻𝗼𝘄𝗳𝗹𝗮𝗸𝗲 𝗔𝗹𝗲𝗿𝘁𝘀! This setup demonstrates how to seamlessly ingest, transform, and monitor data, ensuring immediate notification for any processing failures (like a tricky division by zero error 😉). 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 𝗔𝗴𝗲𝗻𝗱𝗮 & 𝗚𝗼𝗮𝗹: ✅𝗚𝗼𝗮𝗹: Achieve near real-time data transformation and reporting with minimal maintenance overhead. ⏱️ ✅𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: Leverage Snowflake Dynamic Tables (DTs) to declaratively define the entire transformation flow. The pipeline automatically manages dependencies and refreshes complex logic (like filtering for the latest customer record and highest product price) every minute. ✅𝗥𝗲𝘀𝗶𝗹𝗶𝗲𝗻𝗰𝗲: Integrated Snowflake Alerts to actively monitor the DT refresh history. If a data quality issue (like a Division by Zero error 🚫) causes a failure, an email notification is sent immediately 📧, turning passive monitoring into proactive incident response. This project is a great showcase for Snowflake's capabilities in modern, declarative data engineering. Check out the full code and setup on my GitHub! 👇 https://lnkd.in/gq_eH-9U #DataEngineering #Snowflake #DynamicTables #RealTimeAnalytics #Automation #CloudData

#Snowflake data professionals, let's talk pipelines. The declarative vs. manual debate for Snowflake data pipelines. Dynamic Tables: The automated, declarative route. Just set the target state. Streams & Tasks: Full control, but requires manual orchestration. Which approach is the right one, and when?👇 Snowflake #Snowflake #DataEngineering #DynamicTables #StreamsAndTasks

Of course, I'll comment... Three thoughts on the Fivetran dbt Labs merger: Was this on the MDS 2025 bingo card? No, but here's what I think matters beyond the headlines: 🏴 This isn't about "bundling vs unbundling" it's about operational efficiency. The modern data stack gave us flexibility. It also gave us: 12 vendor relationships to manage 12 security reviews to complete 12 contracts to negotiate 12 support channels when things break For a 10-person data team? That's unsustainable. Consolidation isn't ideological. It's practical. And it's frankly overdue. 🏴 The real winners here aren't Fivetran or dbt Labs, they're Databricks and Snowflake. This creates an interesting power dynamic with the data warehouses. Snowflake and Databricks have been building native ingestion and transformation capabilities for years; they want to own the full stack. A merged Fivetran-dbt is now a $600M counterweight sitting upstream of the warehouse layer. That's either: → Healthy competition that keeps warehouse vendors honest OR → A future negotiation battle over who controls the data pipeline Probably both, depending on the quarter. 🏴 But there's a gap no one's addressing: cross-platform visibility. Here's the thing: whether you go with Fivetran-dbt, Databricks, or Snowflake... Your data quality problems don't respect vendor boundaries. A schema change in Salesforce breaks a dbt model, which breaks a Tableau dashboard, which breaks a Monday morning executive meeting 🧑🚒 The platforms will tell you WHERE things broke in their system. They won't tell you WHY it broke three steps upstream or WHAT breaks downstream. That's not a knock on platforms; it's just not their job. Their job is to run infrastructure reliably. Your job is to ensure data is trustworthy across ALL the infrastructure. Those are different problems requiring different solutions. #dataengineering #dataquality #fivetrandbt

🚀 Introducing Snowflake Optima: Effortless Performance at Scale Unlock unprecedented speed with Snowflake Optima - the next-gen optimization engine that automatically accelerates your workloads without extra cost or configuration! ✨ Key highlights: • Optima Indexing: Auto-creates search indexes for recurring queries • 15x faster query performance • 96% micro-partition pruning efficiency • Zero manual tuning required Performance meets simplicity - let your data work smarter! #Snowflake #DataEngineering #CloudData #SnowflakeOptima #SnowflakeSimplifiedbyPrafulkumarSingh Snowflake Snowflake Developers Snowflake Challenge Snowflake Public Sector

Being part of an organization that has been a Snowflake Elite Partner has given me deep exposure to the Snowflake ecosystem -- and one key realization that is something crucial: technical expertise alone isn't enough when you're customer-facing. Earning the Snowflake Sales Professional Certification helped me strengthen the business side of that understanding - learning how to translate pain points into solutions -- a critical skill for any data engineer working directly with clients. Here's what I can now bring to conversations: End-to-end knowledge that drives business value: - Opening new revenue channels through data monetization. - Optimizing business processes and reducing technical debt. - Strengthening risk, regulatory, and compliance postures. Unique capabilities of Snowflake that I didn't know before this certification: - Global Data Clean Rooms - Secure, privacy-preserving collaboration across regions and clouds without moving data. - Unistore – unified OLTP and OLAP on a single platform via hybrid tables (HTAP), eliminating traditional database silos. The bigger picture: Understanding different personas matters. Application developers and software engineers aren't just database users—they're internal champions and evangelists. Knowing how to engage them amplifies our impact. Bottom line: In a customer-facing role, bridging the gap between technical capabilities and business outcomes isn't optional -- it's essential. #Snowflake #DataEngineering #SalesEngineering #ContinuousLearning #PartnerSales #ChannelSales #TechnicalSales https://lnkd.in/gYV_MsWu

✅ 𝗦𝗻𝗼𝘄𝗳𝗹𝗮𝗸𝗲 𝗢𝗽𝘁𝗶𝗺𝗮 is here, and it’s a complete game-changer for query performance. If you're still manually tuning your workloads, you're falling behind. ⁉️ Why is this a big deal? For too long, data teams have been stuck in a reactive cycle: ❌ Manual Tuning: Wasting hours sifting through query profiles to find bottlenecks. ❌ Constant Adjustments: Endlessly tweaking warehouses and clustering keys for small performance gains. ❌ A Never-Ending Battle: Trying to keep up with ever-changing data and query patterns. What if there was a smarter, more automated way? ✅ Introducing Snowflake Optima: an intelligent optimization service that works behind the scenes to make your queries faster and more efficient, with no manual effort required. ⚡ Think of it as an 𝗲𝘅𝗽𝗲𝗿𝘁 𝗱𝗮𝘁𝗮 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿, 𝗯𝘂𝗶𝗹𝘁 𝗿𝗶𝗴𝗵𝘁 𝗶𝗻𝘁𝗼 𝗦𝗻𝗼𝘄𝗳𝗹𝗮𝗸𝗲, 𝘁𝗵𝗮𝘁'𝘀 𝗮𝗹𝘄𝗮𝘆𝘀 𝗼𝗻, 𝗮𝗹𝘄𝗮𝘆𝘀 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴, 𝗮𝗻𝗱 𝗮𝗹𝘄𝗮𝘆𝘀 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗶𝗻𝗴. Here’s how it changes the game: ➡️ 𝗖𝗼𝗻𝘁𝗶𝗻𝘂𝗼𝘂𝘀𝗹𝘆 𝗔𝗻𝗮𝗹𝘆𝘇𝗲𝘀: It learns from your specific workload patterns to identify performance opportunities. ➡️ 𝗔𝗽𝗽𝗹𝗶𝗲𝘀 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝘁 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝘀: It automatically creates and maintains necessary indexes and other optimizations behind the scenes. ➡️ 𝗕𝗼𝗼𝘀𝘁𝘀 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆: By minimizing data scanning and optimizing query execution, it drives faster performance and better resource utilization. ➡️ 𝗭𝗲𝗿𝗼 𝗖𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝘁𝗶𝗼𝗻: It’s on by default. You get the benefits without any setup or ongoing maintenance. 𝗧𝗵𝗲 𝗸𝗲𝘆 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆: Snowflake Optima brings proactive, intelligent optimization directly to your data. It turns performance tuning from a manual chore into an automated, continuous process. ✅ This is a massive leap forward, freeing up your team to focus on innovation and analysis, not infrastructure. ⁉️ Ready to stop tuning and start accelerating? Learn more about Snowflake Optima in the official blog post👇 Shreya Alejandro Stephen Chris Carl Osama #Snowflake #DataEngineering #Performance #Optimization #CloudData #DataWarehouse #Automation

Snowflake continues to innovate with Optima. One of the most exciting features is automatic and proactive search indexing of frequently used point-lookup queries. More search indexing = faster performance. This all occurs at no additional cost!

𝗦𝗻𝗼𝘄𝗳𝗹𝗮𝗸𝗲 𝗗𝗮𝘁𝗮 𝗟𝗼𝗮𝗱𝗶𝗻𝗴 — 𝗘𝗻𝗱-𝘁𝗼-𝗘𝗻𝗱 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 I recently worked on a hands-on project exploring different ways to load data into Snowflake, covering both manual and automated ingestion techniques. This project demonstrates how Snowflake empowers developers and analysts to move, manage, and transform data efficiently across multiple sources. Here are three practical patterns teams commonly implement, mapped to UI, CLI, and cloud workflows alike. 𝗗𝗶𝗿𝗲𝗰𝘁 𝗰𝗿𝗲𝗱𝗲𝗻𝘁𝗶𝗮𝗹𝘀 (𝗦𝟯 𝗮𝗰𝗰𝗲𝘀𝘀 𝗸𝗲𝘆𝘀): Create a stage that embeds cloud storage credentials, then load via COPY using a reusable file format for CSV/JSON/Parquet parsing. Simple to start and good for prototypes, but rotates keys more often and centralizes secrets in Snowflake metadata, so use tight policies and short-lived keys. 𝗦𝘁𝗼𝗿𝗮𝗴𝗲 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 (𝗺𝗮𝗻𝘂𝗮𝗹 𝗖𝗢𝗣𝗬): Use a cloud role and a Snowflake storage integration to establish a trust relationship, then point a stage at the bucket/prefix and run COPY when desired. This removes embedded secrets, delegates auth to the cloud IAM role, and standardizes least-privilege access to specific buckets and paths. 𝗦𝘁𝗼𝗿𝗮𝗴𝗲 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 + 𝗦𝗻𝗼𝘄𝗽𝗶𝗽𝗲 (𝗮𝘂𝘁𝗼-𝗶𝗻𝗴𝗲𝘀𝘁): Keep the storage integration and stage, then add a pipe that triggers COPY automatically as new files land via cloud notifications. Best for continuous ingestion with low-latency updates; operations shift to monitoring pipes, notifications, and load history rather than scheduling COPY jobs. 𝗧𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝗶𝗲𝘀 & 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀: Snowflake Database, SnowSQL CLI, Cloud Storage (S3), Stages, Storage Integrations, Snowpipe, and Time Travel. 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗢𝘂𝘁𝗰𝗼𝗺𝗲: This project deepened my understanding of Snowflake’s data ingestion architecture, secure cloud integrations, and automation capabilities — key components for modern data engineering pipelines. If you’re working with Snowflake or cloud data platforms, mastering these loading methods can significantly enhance your data reliability and scalability. 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 𝗥𝗲𝗽𝗼𝘀𝗶𝘁𝗼𝗿𝘆:https://lnkd.in/dWGQK32N #Snowflake #DataEngineering #CloudData #DataIntegration #Snowpipe #AWS #ETL #DataAnalytics

Snowflake runs fast — but not always efficiently. Over the years, I’ve reviewed hundreds of Snowflake environments, and the same performance issues appear again and again. The good news? Most can be fixed with a few practical tweaks. Here are some of the biggest wins: ✅ Use the Query Profile — find your real bottleneck instead of guessing. ✅ Keep queries simple — avoid “views on views.” They hide complexity and kill performance. ✅ Check your GROUP BY — high-cardinality columns can cause hours of sorting and spilling. ✅ Don’t scale everything up — move specific queries to a larger warehouse instead. These aren’t theory — they come from real-world tuning sessions across dozens of production systems. Small adjustments, massive gains. 💡 These are just a few of the many strategies I cover in my Snowflake Masterclass Training Course #Snowflake #DataEngineering #PerformanceTuning #SQLOptimization #CloudData #datasuperhero #snowflake_influencer