10 min read
Finding the joins your warehouse never declared
Statistical foreign-key inference, validated against real data. How ktx builds the join graph agents rely on.
Read moreBlog
Read practical guides, software comparisons, and implementation advice for context layers, conversational analytics, semantic layers, and AI data agents.
Latest posts
Showing 1-12 of 119 posts
Statistical foreign-key inference, validated against real data. How ktx builds the join graph agents rely on.
Read moreAnthropic published how its data team automated close to 100% of business analytics at 95%+ accuracy. The four-layer architecture they describe is the same one we had already open-sourced as ktx, an independent, Apache-2.0 context engine for data agents.
Read moreA context layer puts your warehouse schema, joins, metric definitions, and business knowledge in one reviewable place so data agents query governed context instead of guessing field names. A look at how it works, and at ktx, the open-source context layer.
Read moreFan and chasm traps, and how ktx compiles intent into safe SQL over a reviewed join graph.
Read morektx is the open-source, executable context layer for data agents. It ingests your data stack and internal docs and turns them into an executable semantic layer plus a wiki, so agents query ktx for correct SQL instead of guessing from raw schema.
Read moreA practical guide to data quality gates for AI analytics agents, covering freshness, completeness, uniqueness, relationships, schema changes, semantic checks, and review routing.
Read moreA practical AI analytics audit workflow for data leaders, covering answer evidence, access logs, metric definitions, lineage, review controls, and compliance-ready documentation.
Read moreA data leader playbook for keeping AI analytics answers consistent with BI dashboards by reusing semantic logic, dashboard context, source priority, tests, and review workflows.
Read moreA change-management workflow for AI analytics metric definitions, covering ownership, versioning, impact analysis, regression tests, release notes, and agent synchronization.
Read moreA practical pilot plan for data leaders rolling out AI analytics with bounded scope, governed metrics, real question tests, human review, monitoring, and rollback criteria.
Read moreA build-vs-buy decision framework for data leaders deciding whether to build an AI analytics context layer internally or use a governed platform.
Read moreA practical guide to governing AI agent access to business metrics with roles, row-level security, semantic context, MCP boundaries, and monitoring.
Read more