Data Architect / Engineering Lead

About Rotageek

Rotageek helps businesses optimise staff scheduling through data-driven technology. By leveraging machine learning and advanced algorithms, we enable companies to forecast demand, improve workforce management, and empower employees with greater control over their schedules. Our solutions drive operational efficiency, enhance customer experiences, and reduce staff turnover across the retail and healthcare sectors.

Rotageek is now part of the ELMO Software group, strengthening our ability to deliver industry-leading workforce management solutions as part of a broader, award-winning HR technology ecosystem.

Our values

• Start with People: protects the people whose data the estate holds, PII handled with rigour, and grows the engineers who build on the architecture rather than gatekeeping it.

• Clear is Kind: documents the model, contracts and standards so the path is clear; names modelling and security trade-offs plainly, and gives direct, growth-oriented review feedback.

• Own It, Together: owns the estate holding end-to-end, correctness, security, cost, coordinates across the Engineering, data and geographic boundaries, and fixes the problem rather than assigning it across the contract line.

Summary

A senior, hands-on data role and the functional lead for Data Engineering, you own the analytics data architecture and the standards the discipline builds to. Data modelling is the core of the role. Reports to Engineering leadership, working closely with our global engineering team in Australia (AU) and the wider product function.

The Data Architect / Data Engineering Lead turns an accreted analytics estate into a modelled, governed one. The core is data modelling: you decide the layered warehouse model, the grains, the conformed dimensions and the semantic layer, and hold the line so dimensional logic lives in the warehouse rather than sprawling across the BI tier. You own the architecture end-to-end, layering, contracts with Engineering, multi-tenant security, deployment as code, and you are the bridge from the business question to the model that answers it. As the functional lead for Data Engineering you own the discipline's standards and are its technical authority and voice in cross-functional decisions, working AI-first and designing the semantic layer for trustworthy AI consumption. You lead through standards and mentoring, not by managing people; your seniority shows in an estate that scales without a rebuild and a discipline that builds the next model to your standard without you in the room.

Key Responsibilities

• Own the analytics data architecture end-to-end, layers, modelling conventions, the contract framework and the catalog/semantic standard, and hold the line on what is modelled where.

• Build the foundational curated layer yourself: dimensional models with stated grain that let the business ask cross-subject questions from day one.

• Establish and hold the data contracts with Engineering, driving the coordinated migration when they must evolve.

• Own multi-tenant security and the estate's operational health: isolation, RLS, PII, and the freshness, quality and cost SLOs.

• Make deployment boring: CI/CD for warehouse and semantic layer, automated provisioning, and staging/production parity.

• Lead the Data Engineering function: own its standards, technical and hiring bar and roadmap input, represent data in cross-functional and architecture decisions, and mentor so the architecture is shared knowledge rather than resting on one person.

• Bridge the business question to the model, and keep the estate coherent with the group data function and AU team across data-residency boundaries

Operating expectations (KPIs)

The measurable bar this role is held to:

• Curated models carry a stated grain, owner, description and tests, at the agreed coverage.

• Freshness, volume and schema SLOs met for critical datasets.

• Zero PII / tenant-isolation incidents.

• Model and semantic-layer changes ship via CI/CD, not the console.

• Warehouse cost within budget, with no unreviewed regressions.

• Natural-language analytics accuracy above threshold on the governed semantic layer

Ideal Candidate Profile

How you work

• Thinks in grains, conformed dimensions and semantic layers, and models with intent rather than reflex.

• Defines the approach up front, and can hold the architecture in their head and draw it on a whiteboard.

• Treats data as a production system, with application-grade operational rigour.

• Force-multiplies through standards and self-service, so analysts and AI agents self-serve without the architect in the loop.

• Leads the discipline, not just the task: sets data's technical direction across the group and holds the bar.

• Data- and cost-informed, treating warehouse spend and tenant-isolation leaks as design constraints, not a later hardening pass.

• Business-minded, owning the path from question to model and pushing back on complexity that doesn't serve it.

• Mentors through the work and owns mistakes as team learning, running blameless data-incident post-mortems

Skills

• Data modelling mastery, the core of the role: dimensional modelling (facts, conformed dimensions, stated grain, SCD types) and cross-system entity resolution; sees the handful of well-modelled subjects a sprawling BI layer should resolve to.

• Warehouse architecture: designs the layered model (raw → staged → curated) and the governed semantic layer, keeping dimensional logic in the warehouse, and platform-agnostic enough to survive a warehouse or tooling migration.

• Data contracts: designs the producer–consumer contract with Engineering, schema, semantics, ownership, freshness, so source changes don't silently break downstream.

• Multi-tenant security: isolation and row-/attribute-level security defined high and inherited rather than re-applied per view; PII classified, masked and least-privilege, handled to GDPR with the audit trail an ISO-certified estate demands, with a clear view of where the model could leak.

• Analytics as software: Git-based models, CI/CD for warehouse and semantic layer, tests, environment parity and observability; SLOs, backfills and quarantine as routine, and warehouse cost as a design input.

• AI-first: uses AI to multiply, generating and refactoring models, tests and docs, reverse-engineering schemas, and designs the semantic layer for trustworthy natural-language analytics; treats AI output as a draft to verify wherever PII or a customer-facing number is at stake

What we're looking for

• Deep, hands-on data modelling for analytics, the non-negotiable: dimensional modelling, semantic-layer design, entity resolution, with real ownership of a curated layer others consumed.

• Cloud data warehouse depth, Snowflake a plus, not a gate: strong modellers on BigQuery, Redshift or Databricks transfer.

• A transformation framework (dbt or equivalent), strong SQL, and the judgement for which tier (warehouse / BI / app) logic belongs in.

• Multi-tenant analytics and analytics-as-software, RLS and tenant isolation; Git, CI/CD, tests, observability.

• Sensitive/PII data under GDPR, access control, masking, audit, retention.

• Functional / discipline leadership of a data area, owns the standards, sets the bar and represents data cross-functionally, without necessarily line-managing.

• Bonus: a specific semantic-layer platform; HR/workforce-management domain; off-legacy-BI migration; UK + AU residency