数据架构师

🚢 Discover OneOcean:

OneOcean is a unified brand born from the integration of OneOcean and Ocean Technologies Group.
Owned by Lloyd’s Register, an organisation with more than 260 years of trust, integrity and leadership at sea, OneOcean combines the agility and ambition of a fast-moving innovator with the strength and stability of one of the world’s most trusted maritime institutions.
At the heart of OneOcean is a portfolio unlike any other in maritime. A comprehensive, integrated portfolio built on years of expertise, trusted by thousands of maritime professionals around the world.
🎯 Our Mission: Our mission is clear. In the race to zero emissions, our research, advisory and technical expertise and industry-firsts are supporting a safe, sustainable maritime energy transition. Today we are a leading provider of classification and compliance services to the marine and offshore industries, helping our clients design, construct and operate their assets to accepted levels of safety and environmental compliance. 🔍 Why Join OneOcean Crew?

• Legacy & Innovation: We were created more than 260 years ago as the world’s first marine classification society to improve and set standards for the safety of ships.

• Global Impact: Our digital solutions are relied upon by more than 30,000 vessels, following the acquisition of OneOcean in 2022 and Ocean Technologies Group in 2024.

• Product Offering: Covering five proven product areas - learning, fleet operations, compliance, voyage planning and performance management - supporting its customers from ship to shore, from training and people operations, to voyage compliance and optimisation.

🧭 Navigating the position: Data Architect

As a Data Architect at OneOcean you will own the data architecture across our growing data organisation — partnering with the Data Engineering, BI and AI teams to facilitate roadmap delivery. You will own the discovery, documentation and specification of the data structures and models required for our products and analytics, including reverse-engineering source data from OneOcean SaaS products and helping design the right data stores for each use case. You will be a senior individual contributor working hand-in-glove with the Data Team Lead, BI Team Lead and AI Team Lead — providing the architectural backbone that keeps their delivery aligned, coherent and scalable.

You will work across the full breadth of our data estate — operational sources, lakehouse, analytical serving (Apache Druid, Apache Superset, Power BI) and the vector stores powering our AI products. You will set modelling standards, write the specifications that engineers build from, and ensure that what we build today doesn't paint us into a corner tomorrow.

This is a fully remote, Romania-based role.

🚢 Your Voyage Ahead:

• Own the data architecture across OneOcean's data organisation — operational, lakehouse, analytical-serving and vector layers — providing a coherent target state and a pragmatic path to it.

• Lead the discovery, documentation and specification of the data structures and models required to support product and analytics roadmaps.

• Reverse-engineer source data from OneOcean SaaS products — building accurate logical models of upstream systems and identifying the data contracts, grain and semantics our pipelines depend on.

• Design the right data store for each use case — choosing between OLTP, columnar / OLAP, lakehouse and vector approaches; making the trade-offs explicit.

• Define and maintain conceptual, logical and physical data models; produce clear ERDs, lineage and dimensional designs (star / snowflake, conformed dimensions, surrogate keys).

• Establish and enforce modelling standards, naming conventions, data contracts and schema-evolution practices across teams.

• Partner with the Data Team Lead, BI Team Lead and AI Team Lead — translating product and analytics needs into specifications their engineers can build from, and unblocking architectural decisions as they arise.

• Collaborate with Product, Architecture and Engineering Team Leads to align data direction with the wider engineering strategy.

• Champion data governance — cataloguing, lineage, ownership, quality, security and privacy.

• Document architectural decisions (ADRs) so the why is preserved alongside the what.

• Mentor engineers across the data organisation on modelling, design and architectural reasoning — without line-managing them.

• Stay current with the data landscape and bring in proven techniques as they mature; foster a culture of continuous improvement, innovation and knowledge sharing.

🚢 Recommended to bring on board:

• 8+ years of commercial experience in data engineering, BI engineering or data architecture roles, with significant time spent on modelling and specification.

• Demonstrable experience owning data architecture across multiple workloads — operational, analytical, lakehouse and (ideally) AI / vector.

• Deep data-modelling expertise — conceptual, logical and physical; dimensional / star-schema; SCD; data contracts; semantic-layer design.

• Strong SQL across multiple dialects; comfortable reading and reasoning about complex source-system schemas.

• Proven ability to reverse-engineer SaaS / operational systems — discovering grain, primary keys, relationships, soft-delete patterns and quirks that aren't in the docs.

• Solid working knowledge of analytical serving (e.g. Apache Druid, Apache Superset, Power BI) and the modelling patterns that make them perform.

• Working knowledge of lakehouse approaches (e.g. Delta Lake) — MERGE semantics, partitioning, schema evolution.

• Pragmatic decision-making — able to balance ideal architecture against delivery pressure and explain the trade-offs clearly.

• Excellent written and visual communication — produces specifications and diagrams that engineers can build from with minimal ambiguity.

• Exceptional collaboration and stakeholder skills — comfortable bridging engineering, BI, AI, product and business audiences.

• Documentation discipline — Confluence-grade architectural records, ADRs, model dictionaries and onboarding material.

Nice to Have

• Familiarity with vector databases and the modelling considerations for AI / RAG workloads (chunk strategy, metadata schema, embedding versioning).

• Exposure to Data Vault or Anchor modelling alongside dimensional approaches.

• Catalog / lineage tooling (e.g. DataHub, Atlan, Collibra, or similar) and data-quality frameworks (e.g. Great Expectations, Soda).

• Experience with data contracts as code (e.g. Open Data Contract Standard) or schema registries.

• Awareness of relevant data-protection regimes (GDPR, UK GDPR) and how they shape modelling decisions (PII tagging, retention, minimisation).

• Python familiarity sufficient to read pipeline code and collaborate on PRs without needing translation.

• Cloud-platform awareness — AWS preferred — and IaC familiarity (Terraform).

• Maritime or other regulated-industry domain experience.

Originally posted on Himalayas