ROLE SUMMARY We are looking for a Databricks App Developer to build and run the application layer that brings SQL data from an external source system into a client’s Databricks environment, and then puts that data in front of business users through a Databricks App. This is a build-and-configure role inside someone else’s estate. The client’s platform team owns Unity Catalog governance and issues the grants; you design and deliver within those guardrails. You will own ingestion configuration, catalog and schema design, workspace and warehouse setup, deployment automation, and the app itself. Important scope note: this is not a Spark or Python data engineering position. Spark transformation work is owned by a separate development team. You need enough working familiarity with what those Spark jobs do to integrate cleanly with them — but deep PySpark, Scala, or Python data engineering is not what we are hiring for here.---------A strong candidate can speak to all of the following from direct, hands-on delivery — not from training or observation. Unity Catalog identity and security model, as a builder inside someone else’s estate. Service-principal versus on-behalf-of-user execution; account-level versus workspace identity; designing within grants that a client platform team owns and controls. A shipped web application that executes user-supplied SQL against Databricks SQL warehouses with correct identity handling. Databricks Apps hosting is preferred; the same pattern delivered on another host is acceptable if the identity and warehouse-execution work was theirs. Databricks Asset Bundles across dev / sandbox / prod, plus the Jobs API — including triggering a job from an external event. Databricks workspace and infrastructure administration — SQL warehouses, compute and app configuration, secrets, connections, credentials, entitlements, permissions, and cost monitoring. Schema and data modelling in Unity Catalog — catalog/schema/table design, naming standards, environment separation, and schema evolution. Advanced SQL. Databricks SQL to a high standard, plus enough source-side SQL (T-SQL, Oracle, or similar) to read and reason about the external system being ingested. Ingestion configuration from an external SQL system into Databricks — connectors, incremental loading, scheduling, reconciliation, and failure handling. Consultant posture. Comfortable delivering inside another organisation’s guardrails, escalating access blockers cleanly, and communicating in writing across time zones. 1. The Databricks App Build, configure, deploy, and support a web application hosted on Databricks Apps that serves data to business users. Deliver an app that executes user-supplied or parameterised SQL against Databricks SQL warehouses with correct identity handling end to end. Make and defend the authorisation design: on-behalf-of-user execution versus app service-principal execution — knowing which one a given screen or query needs, and what each implies for row-level access and auditability. Manage app configuration, secrets, resource bindings, permissions, and app lifecycle across environments. Handle front-end delivery for the app (for example Node.js/React or an equivalent supported framework), including query result rendering, parameter input, error handling, and performance under real user load. 2. Ingestion from the External SQL Source Configure and operate ingestion that lands data from an external SQL system into Unity Catalog — using managed connectors (Lakeflow Connect), JDBC/connection objects, Lakehouse Federation, or the pattern the client platform team has standardised on. Set up incremental and full-refresh loads, scheduling, watermarking, reconciliation, and failure handling. Own source connectivity concerns: connection objects, credentials, service principals, network reachability, and coordination with the source-system owners. Validate landed data against the source — row counts, types, nulls, late-arriving records — and document the checks. 3. Unity Catalog, Schema, and Access Design the catalog / schema / table layout for landed and curated data: naming standards, object structure, and where each environment’s data lives. Work fluently inside the Unity Catalog identity and security model as a builder in a client-owned estate: account-level versus workspace-level identity, service principals, groups, grants, ownership, storage credentials, and external locations. Specify the grants your app and pipelines require, request them through the client’s process, and design so nothing depends on privileges the platform team will not give you. Apply schema evolution and change management without breaking downstream consumers. 4. Workspace and Platform Configuration Configure and right-size SQL warehouses for app and reporting workloads; tune for concurrency, cost, and response time. Handle workspace-level administration relevant to the build: compute policies, app compute, secret scopes, connection and credential objects, entitlements, and permissions. Monitor usage, query history, and system tables; keep the solution inside agreed cost envelopes and flag risks early. 5. Deployment and Automation Package and promote the app, jobs, and supporting objects with Databricks Asset Bundles across dev, sandbox, and production, with environment-specific configuration handled properly. Build and operate jobs through the Jobs API, including triggering a job from an external event rather than a schedule. Work in Git, participate in code review, and integrate with the client’s CI/CD pipeline. 6. Operations, Documentation, and Handover Monitor pipelines and the app in production; triage and resolve failures; own the fix through to closure. Tune SQL and warehouse behaviour when queries are slow or expensive. Produce runbooks, architecture notes, and access-model documentation so the client team can operate what you build.