"Use Snowflake Cortex AISQL for dynamic table analysis"

This is wild. I am not sure if that works as a process but can’t clearly identify from my head right now why it shouldn’t. You can run quality checks and do modifications in later steps. This means declarative AI data transformation pipelines are as simple as writing a select statemen. No API integration needed. No separate scheduler needed. No orchestrator needed (dynamic tables determine the correct build order automatically). No separate lineage needed. No vendor for AI needed. No training for developers needed. No separate monitoring needed. All that you need is a CI pipeline and setup staging. But I got a feeling that Snowflake will automate this, too with the new workspace experience in the near future. Want to hear more about how snowflake gets data teams from any level to elite level performance? Visit adesso SE, October 1st at our booth on the snowflake world tour: Register today: https://lnkd.in/e_qwGZU2

Snowflake just rolled out something game-changing: AI SQL within dynamic tables. 📊 This means you can now enrich your data pipelines with AI-powered insights as the data is ingested, without waiting for post-processing. Imagine: ❄️ Instantly classifying messy data with functions like AI_CLASSIFY ❄️ Running sentiment analysis on customer reviews or survey responses in real time ❄️ Unlocking smarter, higher-value transformations than traditional ETL Instead of bolting on AI later, you can now bake intelligence directly into your pipelines. #AI #MachineLearning #SnowflakePartner #GotCortex?

New updates on AI-D-Ants : - Compatible with databricks and ollama ( let me know if you want more AI endpoints ) - Dashboard feature to visuals your data - All calculation are in a table format as output as well - The code is saved as well to double check More features are coming soon Repo : https://lnkd.in/egUGTAdy

Day 2️⃣ of #RunRevealWeek: What makes our approach to detections better? At RunReveal, we think your detection findings aren't that different from your logs. So we started storing them the same way—same database, same tools, (different table). This simple approach enables: ✅ Instant historical queries across months of data ✅ Easy integration with AI agents that actually understand your alerts ✅ The ability to quickly focus on the fields and data that actually matter Read more about how we think about detections, alerts, and the architectural choices we made to support this thinking in our newest blog part of #RunRevealWeek 👇 👇 👇 https://lnkd.in/d265yKq4

I built a secure API + dashboard for my database in 5 minutes. Claude Code wrote the API. Claude Desktop let me chat with the data. DreamFactory MCP made it all secure. This isn’t AI writing SQL. This is AI orchestrating secure APIs. 👉 Deploy with one command: npx @dreamfactory/create-df-mcp 👉 Our DXT file: https://lnkd.in/gd7j7zjj

The DreamFactory platform has always been this super reliable and capable swiss army knife for APIs, but unless you were a pretty sophisticated technologist its ROI has always been a bit difficult to understand. LLMs have not only given DreamFactory a really useful new client interface, but the MCP server has made DreamFactory's ROI instantly recognizable to non-techies (who let's face it are often making critical technology decisions). Generally speaking I think MCPs have *enormous* potential for rethinking product demos, and this is exactly what I'll be talking about at O'Reilly Demo Days next week! - https://lnkd.in/ehuM_A-e

Now in beta: the remote dbt MCP server. Bring structured, governed context into AI workflows, without local setup or servers. One endpoint to your models, metadata, tests, and lineage makes it simple to: • Power conversational analytics with governed metrics • Search metadata instantly with Discovery • Run SQL, tests, and lineage checks remotely Read more here 👉 https://lnkd.in/ePzTynjF

I've fallen deep into MCP rabbit hole lately, shoutout to this youtube playlist by Data Goblins that got me out of the trenches - it shows you how to tap into Fabric Semantic Models via MCP https://lnkd.in/g4GU4RHX The series walks you through everything - spinning up a MCP server - hooking it up with Fabric objects - actually querying semantic models in a clean, extensible way Super useful if you are a MCP noob like me. And this is more important than u think because > MCP is the glue layer for AI integrations > while semantic models power the core of enterprise analytics > put them together, and you get a whole new way of doing self serve analytics If this topic interest you, this series might save you a lot of trial and error just like it did for me.

Databricks introduced the Data Science Agent, turning their Assistant into an autonomous partner for end-to-end analytics workflows. Here is how it works: Toggle Agent Mode, describe your task, and watch it plan and execute multi-step data science workflows autonomously. Link to the blog post in the comments! #databricks #agents #llm —- ♻️Repost to help your network. And follow Sudarshan Koirala for more. And if you want more material on AI Agents, Data Science and LLM in general, here: 👉 https://lnkd.in/ea2AsS-s

Explored n8n’s Active & Inactive workflow modes today—active workflows run automatically on trigger, inactive ones don’t execute unless manually triggered. Plus, learned about n8n’s supported data types: JSON, binary files, tables, and arrays—perfect for flexible automation! #n8nAutomation #WorkflowModes #DataTypes #TechExploration thank you TrueHorizon AI for the great tutorial