Gravity is positioning as a seed horizontal AI infrastructure play, building foundational capabilities around rag (retrieval-augmented generation).
As agentic architectures emerge as the dominant build pattern, Gravity is positioned to benefit from enterprise demand for autonomous workflow solutions. The timing aligns with broader market readiness for AI systems that can execute multi-step tasks without human intervention.
Gravity is a software company that provides AI-powered business research to supply chain management, client retention, and marketing.
A combination of domain expertise (ex-Looker/Google founders), a synced enterprise Knowledge Base that captures definitions/targets/playbooks from existing tools, and agentic LLM connectors that let Orion autonomously generate, publish and deliver grounded analyses and recommendations.
Orion appears to ground generative responses in an external knowledge store and connected data sources. The product emphasizes a synced Knowledge Base and multiple data connectors (data warehouses, BI tools), indicating retrieval of business data/definitions to condition generation (classic RAG).
Accelerates enterprise AI adoption by providing audit trails and source attribution.
Claims of proactive 24/7 operation, autonomous publishing/scheduling of analyses, multi-step exploration (follow-ups, root-cause, recommendations) and the explicit 'agentic AI' label indicate an autonomous agent layer that orchestrates tools, queries, and outbound actions on behalf of users.
Full workflow automation across legal, finance, and operations. Creates new category of "AI employees" that handle complex multi-step tasks.
The product supports conversational access to project metrics and analyses (via Claude/Gemini integrations), which implies NL interfaces that translate user queries into analytical operations (e.g., SQL, analytic pipelines, or query execution). It's likely converting plain-language questions into executable queries/analyses.
Emerging pattern with potential to unlock new application categories.
While not explicitly called a graph, the synced, permissioned Knowledge Base plus role/project scoping and 'definitions/targets/playbooks' suggests a semantic/business-logic layer that maps entities, metrics, and relationships — effectively functioning as a permission-aware knowledge graph or semantic layer for analytics.
Emerging pattern with potential to unlock new application categories.
Gravity builds on Claude, Gemini Enterprise, Gemini, leveraging Anthropic and Google infrastructure. The technical approach emphasizes rag.
Agentic analytics pipelines that run scheduled analyses and produce narrative outputs; plus a connector pattern that allows external LLMs (Claude, Gemini) to call Orion as a tool/backend. No evidence of internal multi-model chains or MoE ensembles.
Experience at Google and Looker; deep data analytics and enterprise BI background; emphasizes identifying gaps and inefficiencies in data analysis processes; positioned to leverage Generative AI in analytics
Previously: Google, Looker
Founders' backgrounds at Looker and Google align with the problem of transforming traditional BI into proactive, AI-enabled analytics. Their enterprise analytics exposure and emphasis on GenAI disruption suggest strong market-fit propulsion, though public specifics about leadership composition and track records are limited in the provided content.
product led
Target: enterprise
hybrid
• DMI Partners testimonial (Kevin Dugan)
Provide proactive AI analytics with data storytelling to drive business action and reduce manual analysis
Gravity operates in a competitive landscape that includes Looker (Google), Tableau (Salesforce), ThoughtSpot.
Differentiation: Gravity positions Orion as an agentic AI analyst that proactively generates and publishes insights, uses a synced Knowledge Base for business logic, and integrates directly with LLMs (Claude, Gemini) for conversational/agentic workflows — focusing on autonomous analysis rather than interactive dashboards and modeling alone.
Differentiation: Gravity emphasizes automated, scheduled 'data stories' and an AI-first analyst that performs continuous root-cause analysis and recommendations, plus embedding a chat/agent layer in products — features aimed at reducing hours spent on manual analysis vs Tableau’s dashboard-driven exploration.
Differentiation: ThoughtSpot focuses on search & interactive insight; Gravity markets an autonomous analyst that proactively monitors data 24/7, publishes actionable analyses and ties insights to a Knowledge Base and playbooks for governed action, plus direct LLM connectors for agentic workflows.
Heavy reliance on the customer's semantic layer (dbt, Looker) plus warehouse connectors (BigQuery, Snowflake, Databricks) to ground natural-language/LLM outputs in curated metric definitions — they lean into existing metadata rather than trying to reconstruct business logic from raw tables.
An explicit product-level Knowledge Base that is 'synced from the tools you already use' and 'governed by the people who own them' — this signals an embedding + RAG layer tied to the customer's canonical metric definitions and playbooks instead of a generic knowledge store.
Positioning LLMs as first-class connectors: 'Access Orion directly from Claude as a connector' and 'Connect Orion in Google Workspace through Gemini Enterprise' — they treat external LLM platforms as integration endpoints and UI surfaces rather than embedding a single in-house model, which creates a model-agnostic runtime.
Agentic/autonomous analysis pipeline: claims of proactive 24/7 analysis, automatic publication/scheduling, and proactive root-cause & recommendation generation indicate a multi-stage agent stack — discovery, hypothesis generation, SQL execution, result validation, narrative generation, action publication — running continuously and at scale.
Productized storytelling pipeline: automated conversion of SQL/query results into 'data stories' that are scheduled and distributed. This requires generative templates plus deterministic validation hooks (to avoid hallucination) and localized business language mapping to ensure narratives align with metric definitions.
If Gravity achieves its technical roadmap, it could become foundational infrastructure for the next generation of AI applications. Success here would accelerate the timeline for downstream companies to build reliable, production-grade AI products. Failure or pivot would signal continued fragmentation in the AI tooling landscape.
“"Generative AI is a once-in-a-lifetime technological disruption"”
“"Orion learns your business, spots what matters, and delivers the analysis your team would take days to produce"”
“"Orion turns every analysis into data stories your team can act on — published, scheduled, and sent automatically"”
“"Access Orion directly from Claude as a connector"”
“"Connect Orion in Google Workspace through Gemini Enterprise"”
“Connector-first LLM interoperability: explicit connectors from popular LLMs (Claude, Gemini Enterprise) to let external LLM interfaces query internal KB and analytics without context switching.”
