Data Architect

Montash GmbH
Ansprechpartner: Remobi Talent
Munich,  Deutschland
100% RemoteFreiberuflichab sofort100% Auslastung
EnergiewirtschaftData ArchitectureData HubData VaultInfosphere DatastageData Access LayerCloud PlatformSnowflakeData LineageTerraform

Beschreibung

Semantic Layer & Data Platform Architect
6 month B2B contract | Fully remote | Remobi
Remobi is hiring a hands-on Semantic Layer & Data Platform Architect to join a global energy technology company’s Digital Finance team. This is a chance to shape a greenfield semantic layer from day one, making finance data understandable and usable by both people and AI agents.
About the project
The foundation is a Snowflake-based finance data platform that has been live for around six months. It follows a federated model, where finance expert teams produce their own data products and a central team keeps an overview of what exists. A core team of 5 to 6 people looks after the foundations: data management, CI/CD and gatekeeping, data quality testing, access management and row-level security.
This role sits in a new innovation team. Its first goal is a context and semantic layer on top of the platform, including:
  • An architecture in Snowflake
  • A chosen framework and standard for semantic definitions and ontology
  • Governed workflows to approve or decline semantic changes
  • Interfaces that serve this information to human users and to AI via MCP
The architecture isn’t fixed yet, so you’ll shape it.
The role
You’ll own the end to end architecture and implementation of the semantic layer, from the Snowflake data model through to the interfaces that consume it.
This is a full-stack, hands-on role, not an advisory one. You’ll turn a broad set of competing semantic and AI concepts into a working solution, ship early increments that show what works and what doesn’t, and take full ownership of delivery. As the team grows, you may lead a small group of engineers.
Key responsibilities
Architecture and design
  • Design the target architecture for the semantic layer within Snowflake, integrated with the existing data platform
  • Define how semantic metadata (business definitions, metrics, entities, relationships, lineage) maps to federated data products owned by different finance teams
  • Decide where and how the ontology is stored and versioned, and how it stays in sync with the underlying data products
Framework and standards selection
  • Evaluate and recommend the right framework for semantic modelling, such as Snowflake Semantic Views, Open Semantic Interchange (OSI), dbt Semantic Layer / MetricFlow, or RDF/OWL-based ontologies
  • Document decisions with clear trade-offs and an architecture decision record (ADR) trail
Interfaces and distribution
  • Design and build interfaces that serve semantic information to human users (catalogue, search, documentation) and to AI agents
  • Expose the semantic layer to LLMs and agents via the Model Context Protocol (MCP), and via APIs where relevant
  • Ensure AI consumers respect existing access controls, including role-based and row-level security
Governance workflows
  • Build workflows for proposing, reviewing, approving or declining changes to semantic definitions
  • Integrate semantic changes into the platform’s existing CI/CD, gatekeeping and data quality testing
Delivery and leadership
  • Own implementation of the kickoff project in an agile / Scrum cadence
  • Deliver early prototypes and POCs that show value quickly, then harden them to production quality
  • Lead and mentor engineers as the team grows, working closely with internal stakeholders, the core platform team and the programme manager
Essential
  • Deep Snowflake architecture and engineering experience at enterprise scale
  • Advanced SQL and data modelling (dimensional, Data Vault or similar)
  • Designing and implementing a semantic or metrics layer in production
  • Business glossaries, entity and relationship modelling, and metadata management
  • Experience exposing data or metadata to LLMs and AI agents
  • Building APIs and services in Python (or similar)
  • Federated data platforms and data products
  • Git-based CI/CD, automated testing and infrastructure as code
  • Role-based access control and row-level security
  • Technical ownership of a delivery from design to production
Desirable
  • Snowflake Horizon, Cortex, Semantic Views; SnowPro Advanced certification
  • dbt Semantic Layer / MetricFlow, Cube, AtScale, LookML, Open Semantic Interchange (OSI)
  • RDF, OWL, SKOS, SPARQL, knowledge graphs, Apache Jena or similar triple stores
  • MCP server development, RAG patterns, LLM frameworks
  • REST / GraphQL and data catalogue integration (Collibra, Alation, Atlan, DataHub)
  • Data Mesh principles and data contracts
  • dbt, Terraform, Azure DevOps or GitHub Actions, data quality tooling (Great Expectations, Soda)
  • Approval and change workflows for metadata, data lineage tooling
  • Finance data (controlling, reporting, planning) in a large enterprise
  • Leading small engineering teams in a Scrum setting

The details
  • 6 month B2B contract
  • Fully remote
  • Start: ASAP

Bewerben