Platform. Integrations

Connect everything.
Migrate nothing.

LazyFox works with your existing data stack, structured and unstructured, wherever it lives, however your stack is evolving. No lock-in. No migration projects. No rearchitecting before you can get started.

Book a Demo → See all integrations
Structured + unstructured — warehouses, documents, PDFs, and NoSQL in one semantic layer·
Swap your LLM without rebuilding your semantic layer·
Read-only connection — no data moved, copied, or migrated·
Databricks, Snowflake, Redshift — connect all three simultaneously·
167% NRR — Series A fintech, querying invoice PDFs in natural language via MongoDB·
dbt documentation feeds directly into LazyFox's structural layer·
Most teams reach meaningful coverage within 30 days·
Structured + unstructured — warehouses, documents, PDFs, and NoSQL in one semantic layer·
Swap your LLM without rebuilding your semantic layer·
Read-only connection — no data moved, copied, or migrated·
Databricks, Snowflake, Redshift — connect all three simultaneously·
167% NRR — Series A fintech, querying invoice PDFs in natural language via MongoDB·
dbt documentation feeds directly into LazyFox's structural layer·
Most teams reach meaningful coverage within 30 days·
Our Approach

Your stack is the right stack. We govern the meaning on top of it.

Enterprise data stacks are never clean. They're the result of years of acquisitions, migrations, vendor decisions, and team preferences. You have systems on Snowflake and systems still on Redshift. Your dbt models are partially documented. Your LLM of choice today might not be your LLM of choice in eighteen months.

LazyFox was built for this reality. We connect to what you have, govern meaning across all of it simultaneously, and make sure that when your stack changes (as it will) your semantic layer doesn't have to be rebuilt from scratch.

Your existing stack
Databricks Snowflake Redshift BigQuery MongoDB dbt PDFs Salesforce Tableau Looker Claude OpenAI GitHub Okta
connects to
LazyFox Semantic Layer — no migration required
Meaning governed across all connected systems simultaneously
Integration Categories

Every layer of your stack.
One governed semantic layer above it.

LazyFox connects across seven integration categories, from the warehouse to the model to the BI tool your business users open every morning.

🗄️
Data Warehouses & Lakehouses
Connect LazyFox to your data storage layer. It indexes schemas, tables, and field-level documentation without moving a byte. Mixed environments (multi-warehouse, mid-migration, hybrid cloud) are supported from day one.
Read-only · No ETL
Databricks Snowflake Amazon Redshift Google BigQuery Azure Synapse Analytics PostgreSQL + more on request
LazyFox reconciles meaning across all connected warehouses simultaneously, so a team that runs Databricks and Redshift in parallel gets one governed semantic layer, not two competing ones.
📄
Structured & Unstructured Data
Most enterprise data isn't sitting neatly in a warehouse. Contracts, invoices, reports, PDFs, and documents hold critical business context, and until now, that context has been invisible to AI queries. LazyFox connects to document and NoSQL sources, indexes their content through the semantic layer, and makes them queryable in natural language alongside your structured data.
New category · Differentiating
MongoDB PDFs & document repositories Invoices & financial documents Contracts + additional sources on request
In practice: a LazyFox customer uses this integration to enable natural language reporting and analytics directly from invoice PDFs and financial documents, without extracting data into a warehouse first. Knowledge workers query document content the same way they query structured metrics, governed by the same semantic layer.
⚙️
Transformation & Semantic Tools
LazyFox sits above your transformation layer, not instead of it. We pull in the documentation and context your engineers have already built, and add the governance layer that makes it trustworthy at scale. If your team has invested in dbt documentation, that investment feeds directly into LazyFox's structural layer.
Builds on existing work
dbt Core dbt Cloud + additional tools on request
Structural descriptions and model documentation from dbt are ingested into LazyFox's structural layer automatically. You don't start from zero.
🤖
AI Models & LLMs
LazyFox is model-agnostic by design. You configure which LLM powers which task, embedding, intent recognition, code completion, response generation, and swap models without rebuilding your semantic layer. Vendor lock-in is an enterprise risk. We treat model selection as configuration, not architecture.
Fully configurable · No lock-in
Anthropic Claude OpenAI GPT-4 / GPT-4o Google Gemini Azure OpenAI Mistral Self-hosted models
On-premises deployments can route all model calls through your own infrastructure or approved enterprise API agreements, no external model traffic required.
📊
BI & Analytics Tools
Governed semantic definitions flow downstream to the tools your business users already live in. Rather than letting each BI tool define its own version of a metric, LazyFox ensures definitions originate from one governed source, and propagate consistently everywhere they're consumed.
Downstream consistency
Tableau Looker Power BI + additional BI connectors on request
When a KPI definition is updated in the Semantic Manager, it flows to all connected BI tools without a manual sync step.
🗂️
Data Catalogs
If you're already using a data catalog, LazyFox complements rather than replaces it. We pull structural metadata from your catalog and layer governed semantic definitions on top, adding the logical and contextual layers your catalog doesn't have.
Complementary · Non-disruptive
Databricks Unity Catalog AWS Glue Data Catalog + additional catalogs on request
Your catalog investment isn't wasted. LazyFox builds the logical and contextual governance layers on top of whatever structural metadata you've already captured.
🔧
Developer & Repo Integrations
For teams that manage semantic definitions close to code, LazyFox integrates with your CI/CD workflow, validating documentation quality at commit time and flagging semantic conflicts before they merge into the main branch.
Shift-left governance
GitHub GitLab Bitbucket on request
LazyFox can check whether code documentation meets semantic standards at push time, before ambiguous definitions land in the knowledge layer.
🔐
Identity & Access Management
Existing permission structures from your identity provider can be mapped directly into LazyFox. You don't rebuild access control from scratch, you extend what you already have into the semantic layer.
SSO · RBAC · SAML
Okta Azure Active Directory / Entra ID SAML 2.0 OAuth 2.0 / OIDC + additional providers on request
User, group, and role-level permissions in your IdP map directly into LazyFox's access control model, controlling which semantic units each role can view, propose, or approve.
Unstructured Data

Natural Language reporting from documents. Without a warehouse.

The data your AI needs isn't only in structured tables. Invoices, contracts, reports, and PDFs contain critical business context, and most enterprises have no governed way to query it alongside their metrics.

LazyFox's MongoDB integration indexes document content through the same semantic layer that governs your structured data. The result: knowledge workers can ask natural language questions across both, and get answers that reflect the same governed definitions, applied consistently.

  • Document content indexed and queryable in natural language
  • Same semantic governance layer for structured and unstructured sources
  • No data extraction into a warehouse before querying is possible
  • PDF invoices, contracts, and financial reports treated as first-class data sources
167%
NRR — a LazyFox customer using MongoDB + PDF integration for financial analytics
📄
Invoice PDFs · Contracts · Reports
Unstructured documents, stored in MongoDB
↓
🗄️
Structured data sources
Warehouses, dbt models, CRM records
↓
⚡
LazyFox Semantic Layer
Indexes both. Reconciles meaning across both. Governs both through the same definitions.
↓
💬
"What's our total invoiced revenue for Q2?"
Answered from documents and structured data, governed, consistent, no hallucination
How Connection Works

Connected in hours.
Not months.

LazyFox connects to your data systems via API and native connectors, no agent to deploy, no schema migration, no requirement to centralize data first.

01

Connect your sources

Point LazyFox at your data systems via read-only API connectors. Warehouses, document stores, dbt models, BI tools, whatever's in your stack. No data leaves its home.

Read-only · Hours not weeks
02

Index the structural layer

LazyFox automatically indexes schemas, tables, field documentation, and document content from all connected sources. Existing dbt and catalog documentation is pulled in directly, you don't start from zero.

Automatic · Non-disruptive
03

Build the semantic layer

Your team begins defining KPIs, business rules, and domain context in the Semantic Manager, starting with your highest-value metrics. Most teams reach meaningful coverage within 30 days.

Weeks not quarters

Don't see your stack?

We're adding connectors continuously and prioritize development based on customer need. Most net-new connectors can be scoped as part of enterprise onboarding.

Talk to an engineer →

Your stack is ready.
Is your semantic layer?

We'll map your current stack against LazyFox's connectors and show you exactly what a 30-day path to governed meaning looks like for your environment.

Request a Demo