Looker

Agentic Semantic Layer for AI Agents

An agentic semantic layer is a metadata layer between AI agents and a data warehouse that defines governed metric definitions, enforces access control, and exposes programmatic query interfaces (e.g., MCP or REST/SDKs). Unlike legacy semantic layers built for BI dashboards, an agentic semantic layer is designed for programmatic consumers (LLMs, AI agents, SDKs) and requires machine-readable metric definitions, programmatic discovery/querying, structural multi-tenancy, pre-aggregation to handle high query volume, and schema-as-code with version control. It generates SQL from governed definitions (not from LLMs), scopes queries per-tenant via publishable keys, and returns auditable, consistent results. The article compares existing tools (Cube, dbt MetricFlow, Looker, AtScale, ThoughtSpot, Bonnard) and explains how the agentic layer fits into the modern data stack on top of ingestion, warehouses, and dbt transformations.

Data Workers Chooses MCP for AI-Agent Integrations

Data Workers explains why it adopted the Model Context Protocol (MCP) as the standard interface for its AI agents to connect with modern data-stack tools. MCP, originally developed by Anthropic, provides a universal protocol enabling agents to call external tools if those tools implement an MCP server. The post highlights rapid prototyping, composability between agents, and community‑provided MCP servers as benefits. It also outlines operational challenges the team is addressing: authentication at scale, added latency from multiple tool calls, uneven quality of community MCP servers, handling stateful workflows on top of MCP's request-response model, and increased security surface area. Data Workers is building custom MCP servers for its agents and a context layer to manage stateful data engineering workflows.

Hands‑On Apache Iceberg on Dremio Cloud

This technical walkthrough (Part 14 of a 15-part Apache Iceberg masterclass) demonstrates how to use Apache Iceberg with Dremio Cloud. It covers getting started (creating a Dremio Cloud account and connecting object storage), creating Iceberg tables with hidden partitioning, ingesting data via COPY INTO or INSERT...SELECT, and Dremio platform features such as the Open Catalog (Polaris-based), Columnar Cloud Cache (C3), query federation, semantic layer, Reflections for query acceleration, table optimization and time travel. The article also describes governance features (column- and row-level controls), the MCP Server (Model Context Protocol) to let external LLM agents query governed data, and how Dremio exposes Iceberg to BI tools via ODBC/JDBC/Arrow Flight.