LoginSign Up
← Back to all sparks
O

Omni

ANALYTICS
Velocity5.0

Business intelligence and embedded analytics platform

omni.co ↗

Omni ships AI-native BI at weekly cadence: semantic model generation, agent workbook editing, and MCP tooling all live

business-intelligenceai-nativeembedded-analyticssemantic-modelingmcp-tooling
◆Current state
Omni is a modern BI platform shipping at high weekly velocity, with AI deeply embedded across the stack: an Omni Agent that edits dashboards and workbooks, semantic model generation powered by AI, AI Routines triggered from Slack, and MCP tools for external agents. Embedded analytics went generally available, and the CLI reached v1.2. Korean localization signals international push.
◆Where it's heading
The platform is converging on a model where data analysts interact with BI through conversation and agents rather than through click-by-click dashboard building. Semantic model generation removes the hardest setup step; AI Routines and Slack integration bring analytics into existing team workflows. Expect AI credit governance and embed entity controls to mature as enterprise deployments grow.
◆Prediction
The Omni Agent gaining memory support and workbook editing suggests a path toward fully autonomous dashboard creation from a natural language brief — the roadmap inflection will be when the agent can build a net-new dashboard end-to-end without human configuration.

◆Recent moves

  1. 3d ago

    Apps GA, MCP PNG exports, and custom calendar support

    Dashboard PNG exports from MCP clients let AI agents hand static chart snapshots to external systems, while Apps reaching GA formalizes the embedded analytics path that was in preview — a meaningful milestone for Omni's embed story.

    View source ↗
  2. 11d ago

    Omni Agent workbook edits, AI chat user memory, and Korean support

    User memory in AI chat means Omni retains analyst preferences and context across sessions — reducing repetitive re-prompting and making the AI assistant more practical for daily use rather than one-off queries.

    View source ↗
  3. 19d ago

    Embedded apps, Databricks query tagging, and model forecasting controls

    Embedded apps arriving (prior to formal GA) gave enterprise customers an internal-facing dashboard deployment path; Databricks query tagging improves cost attribution for teams running heavy workloads on that warehouse.

    View source ↗
  4. 25d ago

    Default filters, dashboard query stopping, and full-screen preview

    Default filters on composite topics and dashboard query stopping are quality-of-life improvements for power users — neither changes the product's direction but reduces day-to-day friction for complex data models.

    View source ↗
  5. 1mo ago

    Full-screen presentation mode, searchDashboards MCP tool, and delivery personalization

    The searchDashboards MCP tool gives external AI agents a search interface into Omni's content — part of the same MCP build-out that makes Omni accessible as an agent-callable analytics layer rather than just a UI tool.

    View source ↗
  6. 1mo ago

    AI credit controls for embed groups, Azure AI Evals, and map legend positioning

    AI credit controls per embed entity group let enterprise customers budget AI usage at a granular level — necessary infrastructure for embedding Omni's AI features in multi-tenant products where each customer should pay for their own credits.

    View source ↗