Описание:
EPAM delivers data products on an enterprise Intelligence Platform built on Databricks. The platform includes the Data Landscape Explorer, a register of domains, entities, owners and data products that serves as the main entry point for business users.
Задачи:
- Collaborate with use case owners, domain data owners and stewards to translate their needs into specs and run demo-and-feedback loops
- Own the BMAD spec chain, from product brief and PRD through architecture and stories with acceptance criteria
- Ensure code is written only after spec approval and every merge request traces back to a story and its acceptance tests
- Deliver glossary features in the DLE, including domain-based term search and browsing, and term pages with definitions, synonyms, owners, stewards, status and version
- Implement links from terms to entities and contract fields using the existing `semantic_term_urn` field
- Design an impact view showing which products and consumers depend on a term
- Build a propose-or-change flow that routes requests to steward review and then into Git as a merge request
- Develop the UI using Databricks Apps, governed by Unity Catalog for comments, tags, permissions and lineage, and deploy via Asset Bundles
- Embed security into the design, ensuring authentication fails closed, on-behalf-of auth never silently falls back to a service principal, and access is governed by Unity Catalog
- Operate within the client’s VDI and Coder workspaces, keeping client data inside the secured perimeter
- Record decisions as ADRs in the platform’s knowledge base
- Hand over a working demo, a runbook and a backlog for evolving the prototype into a product
Требования:
- 5+ Years of experience as a Forward Deployed Engineer, Data Engineer or Full-Stack Engineer delivering data products in close collaboration with business stakeholders
- Expertise in Databricks, including Databricks Apps, Unity Catalog and Asset Bundles
- Proficiency in spec-first delivery methodologies such as BMAD, including product briefs, PRDs, architecture documents and stories with acceptance criteria
- Background in building data catalog, glossary or metadata management solutions, with familiarity with DataHub as a stewardship tool
- Skills in designing and implementing data contracts and governed release flows through Git
- Understanding of semantic models, business glossaries and lineage concepts, including linking terms to entities and contract fields
- Competency in secure authentication patterns, including on-behalf-of auth and fail-closed design
- Ability to work with AI coding agents such as Claude to draft, build and test against approved specs, and direct and review their output
- Ability to record architectural decisions as ADRs and produce runbooks and handover documentation
- Flexibility to use restricted development environments such as VDI and Coder workspaces while respecting data perimeter constraints
- Strong communication skills for running demo-and-feedback loops with use case owners, domain data owners and stewards
- Excellent written and spoken English (B2+)
Условия:
Remote hiring locations: Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Uzbekistan