Jedify + Snowflake
Snowflake semantic layer for AI agents
Jedify continuously builds business meaning into governed Snowflake Semantic Views, so the agent’s answers stay accurate and current, even as your business changes.
Jedify + Snowflake: a trusted source of truth
Snowflake delivers Cortex Analyst’s natural-language power. Jedify adds the business meaning behind it: customer segments, pricing logic, revenue terms – so answers stay grounded in how your business actually works. Together with Jedify, Semantic Views becomes a trusted source of truth for business data.
Answers that stay accurate
Domain-scoped Semantic Views keep Agents grounded in business meaning that stays current as your business changes.
Oversight, not just automation
Changes apply automatically or route for approval first, so nothing changes without a HITL.
Compute stays on Snowflake
All context lives in governed Semantic Views, with data staying at source and nothing moving.
Get answers your business can trust, governed the way you expect
1
Connect sources
Jedify connects to Snowflake and external systems (CRM, billing, ops)
2
Build Context Graph
Concepts, metrics, synonyms, fiscal rules, and join relationships mapped
3
Auto-scope
Graph analysis splits the context graph into analytical domains automatically
4
Deploy Semantic Views
Semantic Views created in Snowflake, with verified queries and custom instructions included
5
Query naturally
Queries are answered from day one using verified queries and rich context
6
Stay in sync
Semantic and schematic changes are detected and propagated to Semantic Views
Builds itself, continuously
The Context Graph learns from how your business is queried, then auto-generates and updates Semantic Views on its own – no hand-written semantic model, SQL, or manual rebuild when data changes.
Semantic View updatedREVENUE_ANALYTICS · 2 metrics added · no SQL written
SyncedPulls from everywhere
Business meaning comes from external data sources like CRM, billing, and operational systems, and not just what’s already structured in the warehouse.
KnowledgeBusiness systems
Built to fit Semantic Views
Large, connected context gets scoped into multiple domain-specific views. When the business grows, the views update as well, automatically.
Schema
Monthly Feature Adoption Rates
JanFebMarAprMayJunJul
WITH subscription_monthly_metrics AS
(SELECT MONTH,
organization_id,
tier,
product_plan_name,
start_date, end_date,
is_free_trial
FROM billing.subscriptions)