Updated September 29, 2026
Microsoft’s About-Face on Semantic Layer StandardsWhen Snowflake launched the Open Semantic Interchange in September 2025, dozens of companies signed on. Microsoft wasn't one of them. Power BI's semantic models stayed in DAX, built for Power BI, while the rest of the industry worked on a…
Updated September 22, 2026
A Wrong Answer Has Never Looked So GoodExecutive Summary:An agent doesn’t have a gut check, so it treats a wrong answer as fact and can’t catch it before building the next question on top of it, and accuracy drags down fast as a conversation goes on. The…
Updated September 17, 2026
The Missing Step in Text-to-SQLExecutive Summary: I asked how many units we sold in Q2 2008 against an AdventureWorks tutorial dataset and got 25,427 units. Correct. A second calendar in the same schema, a retail 4-4-5, returned 24,639. Also correct. Claude with direct database…
Updated September 15, 2026
Semantic Layers Are Still the Missing Piece in Generative AnalyticsAsk your AI what churn was last quarter, and it will give you a number. It won't tell you whether that's logo churn or revenue churn, or whether it includes voluntary cancellations. A human analyst filters all of that without…
Updated August 26, 2026
Why AtScale Is Thriving: AI Can’t Do MathIn October 2025, University of Chicago researchers confirmed that the best AI can't multiply two four-digit numbers, a problem fourth graders do with a pencil. The AI models they tested were wrong a whopping 99% of the time. Strange as…
Updated August 21, 2026
The Missing Third Leg of Every Agentic AI StackEvery architecture diagram I see draws the same three boxes. Sometimes it's Model + RAG + MCP. Sometimes it's Model + MCP + Tools. The labels move around, but the shape never changes. The diagram describes how an agent reaches…
Updated August 20, 2026
The Multi-Cloud Metrics Problem Snowflake, Databricks, and BigQuery Can’t Solve AloneEnterprises running Snowflake, Databricks, and BigQuery side by side no longer struggle with data movement. Data pipeline tools solved that years ago. Now they have a metrics problem: the same "gross margin" or "active users" number means something different depending…
Updated August 20, 2026
Inside a Real Benchmark: Measuring the Cost of AI on the WarehouseAny AI performance conversation runs the same three tracks: token costs, inference latency, and model accuracy, and all three numbers matter. But once an AI system reaches enterprise data, a fourth system starts determining both cost and performance: the data…
Updated August 19, 2026
The Other Half of Intelligence EfficiencyAI usage isn't the same as AI value. That’s why Snowflake introduced “intelligence efficiency” this week. In a post on Snowflake's blog, CEO Sridhar Ramaswamy laid out how companies are pouring money into AI, and boards want proof it's translating…
Updated August 19, 2026
Anatomy of a Vague Natural Language Prompt: What an Unguided LLM Does to Your Warehouse"What was revenue in the Northeast region last quarter?" At first blush, the query looks simple enough. It reads like something you'd ask an analyst on your team and get an answer back in five minutes. But an LLM can't…
