Quest Software announced Quest Lakebase Modeler on November 19, 2026, making the schema modeling tool available through Databricks Marketplace. Quest Lakebase Modeler helps developers, data engineers, and architects design, govern, and deploy application-ready schemas directly within Databricks Lakebase. Databricks Lakebase is a fully managed, serverless Postgres database built for AI applications and agents on the Databricks Data + AI Platform.
Why Governed Data Models Matter for AI Applications
As organizations deploy AI agents that read and write operational records, underlying data structures directly determine output reliability. Gartner’s June 2026 research summary “How Master Data Management Boosts the Context Layer for AI Agents” says that AI initiatives fail when agents lack reliable master data context, which amplifies underlying data quality issues.
Data preparation and modeling practices are shifting to keep pace with unstructured and transactional workloads. In its February 2025 article “Four Key Pillars of a Data-Centric Approach to AI,” Gartner says that AI systems increasingly depend on unstructured inputs such as text, images, and video, requiring data architectures to evolve beyond traditional structured-only ETL tools. Gartner recommends a data-centric AI approach built on four pillars (exploratory data analysis and preparation, feature engineering, data labeling and annotation, and data augmentation) and projects that synthetic data will surpass real data in AI models by 2030.
How Quest Lakebase Modeler Builds and Governs Schemas
Quest Lakebase Modeler operates inside the Databricks workspace alongside Unity Catalog governance controls so teams do not leave the platform to manage schemas. Quest Software says early schema choices involving keys, relationships, constraints, and governance policies affect application performance, scalability, and reliability on Lakebase. Stephen Orban, SVP Product Partnerships and Ecosystem at Databricks, said: “We built Lakebase to give developers a Postgres database for these applications. With Quest Lakebase Modeler available on Databricks Marketplace, we’re expanding the partner ecosystem that helps customers build AI applications and agents on trusted, governed data, so customers can design and govern their schemas directly in the Databricks workflow, before their apps and agents start writing to the database.”
Multi-Source Schema Creation and Workload Detection
Through AI-assisted modeling, users can create and refine data models, reverse-engineer existing database environments, map business concepts to data assets, and generate deployment-ready database definitions. Users can start from a plain-language application description, reverse-engineer a live Lakebase database, import Unity Catalog tables, or build a model manually, while the AI assistant detects whether a workload is transactional or analytical.
Governance, Approval Workflows, and Copy-on-Write Branching
A core design principle in Quest Lakebase Modeler separates AI assistance from production database changes. While AI assists in generating models, identifying relationships, and proposing improvements, human experts remain responsible for approving schema changes before deployment. According to Quest Software’s Lakebase Modeler datasheet, AI enrichment proposes relationships, domains, and descriptions in a separate review workbench showing confidence levels and reasoning, where users approve, reject, or edit each item before anything enters the model.
DDL Generation and Branch Validation
Quest Software says maintaining the data model as the authoritative source of truth reduces drift between intended designs and deployed database systems. To enforce that consistency, DDL code is generated directly from the model rather than hand-written, and teams review entities, constraints, indexes, and DDL before promoting schemas across Development, Staging, and Production. Teams can also deploy a schema model to its own Lakebase branch using native copy-on-write branching to test an application against a production-fidelity copy and discard the branch if the design fails.
How Teams Use Quest Lakebase Modeler Across the Lifecycle
Quest Lakebase Modeler supports designing data architectures for new AI applications and agents, documenting existing environments, standardizing practices across teams, and managing application evolution over time. On the product datasheet, Quest outlines specific benefits for four roles: data engineers and application developers, application owners, DBAs and data modelers, and platform and architecture leads. Doug MacWilliams, Databricks MVP and Partner in West Monroe’s Data and Analytics team, said: “Lakebase puts transactional apps directly on the lakehouse, so modeling matters more than ever. Quest has been a leader in modeling tools for decades. Having that built for Databricks gives our teams a solid starting point on every Lakebase build.”
Four Core Lakebase Modeling Workflows
Quest Software lists four primary workflows in its product documentation for application and database teams:
- Building a new application on Lakebase: Describe the application in plain language, review the drafted schema, refine entities, and deploy with documented design decisions.
- Serving Lakehouse data into an application: Model operational tables alongside synced read tables, including the primary keys required for incremental sync.
- Documenting an undocumented database: Reverse-engineer a live Lakebase schema to generate proposed descriptions, domains, and missing relationships for human review.
- Changing a schema in production: Deploy schema updates to a branch, test the running application against it, and promote changes only after validation passes.
Also Read: Saviynt Launches EU Sovereign Identity Security Cloud on S3NS
Quest Software’s Broader Databricks Data Lifecycle Stack
Quest Software serves more than 28,000 companies worldwide, including over 90% of the Fortune 500, across data management and governance, cybersecurity, and platform modernization. For more than 20 years, Quest has supplied database modeling, replication, monitoring, and query tools for Oracle, SQL Server, and PostgreSQL environments, and Quest Lakebase Modeler extends the enterprise data modeling practices of Quest’s erwin software into the Databricks environment.
As detailed in Quest Software’s Databricks solution brief, the company connects to Databricks across five data lifecycle stages using Quest and erwin Data Modeler for design, SharePlex for near-zero-downtime replication, Foglight and Toad Data Studio for performance operations, Quest Data Intelligence for lineage and metadata governance, and Toad Data Point with Databricks Genie for self-service analytics. Rakesh Shah, Vice President of Product at Quest Software, said: “Quest Lakebase Modeler brings decades of enterprise data modeling expertise into the Databricks Lakebase environment, helping teams establish scalable data architectures before the first application reaches production. We’re helping customers move faster while maintaining the governance, consistency and confidence required to deliver trusted business outcomes.”
Questions about Quest Lakebase Modeler
Where does Quest Lakebase Modeler run and store customer data?
Quest Lakebase Modeler installs from Databricks Marketplace and runs inside the customer’s Databricks tenant so data stays in the tenant with no external infrastructure required. Built-in workflow governance separates the person who designs a schema from the person who deploys it, records an audit trail for model changes, reviewed AI suggestions, and deployments, and registers deployed objects in Unity Catalog.
How does Quest Lakebase Modeler map schemas to Unity Catalog?
Source Mapping binds model entities and attributes to existing Unity Catalog tables and columns with automatic or manual matching. For entities that do not yet exist in the catalog, Quest Lakebase Modeler generates new Unity Catalog table structures directly from the model.
What is Databricks Lakebase and how do teams use it for AI agents?
Databricks announced the general availability of Lakebase on February 2, 2026 to unify transactional (OLTP), analytical, and AI workloads on one platform without moving data between separate OLTP databases and the lakehouse. Built as PostgreSQL-compatible OLTP governed by Unity Catalog and reachable through standard Postgres drivers and clients, Lakebase is used by organizations to replace external OLTP systems, eliminate ETL pipelines by syncing operational data with Delta tables, and store agent state, memory, and configuration for stateful AI agents, according to Databricks’ Lakebase launch partner announcement.
































