Insights
Gain expert strategies, real-world perspectives, and best practices in data engineering, analytics, and AI. Explore our latest insights to stay ahead in the ever-evolving data landscape.
Blog posts
Beyond Simple Automation: Building Safe Enterprise AI with Harnesses and Loops
Deploying autonomous AI agents directly onto enterprise ERPs and CRMs introduces security and operational risks. Discover how pairing dynamic agent loops with governance harnesses enables reliable, auditable, and adaptive AI workflows without sacrificing control.
AI Access Management: Three Governance Layers
AI introduces a new access management challenge: what the user can see, what the agent can see, and what the LLM can see are not the same thing. This article explores how AI governance can be integrated into an existing hub-and-domain data architecture through identity groups, secured schemas, and governed AI harnesses.
Hub & Domains: A Practical Data Operating Model
Domains bring data ownership closer to the business, but governance, access management, and shared standards remain difficult to decentralize. This article explores a Hub & Domains operating model that combines business ownership with centralized governance and platform controls.
Certified Metrics: From Fact to Dashboard
Certified metrics require more than documented formulas. Learn how facts, measures, dimensions, aggregate metrics, and dashboards work together to create trusted and reusable business metrics.
Distributing Facts and Dimensions: Governance, Access & Ownership
Building facts and dimensions is only part of the challenge. This article explores how certified data should be distributed across the organization through controlled access paths, ownership models, governance processes, and support structures.
What the Claude Code Leak Reveals About Enterprise AI
The Claude Code leak provides a practical look at AI architecture and where value is created in enterprise AI applications. This article explores why governance, workflows, and deterministic logic often matter more than AI itself when building reliable, cost-effective solutions.