Job Description
Responsible for the design, development, and maintenance of the data pipelines, models, and architecture that move data from its source through to a marketing semantic layer, and for ensuring that data is structured, reliable, and ready for use. This role is the accountable technical owner of that semantic layer, built on large centralized semantic models that sit on top of the raw data tables and turn them into a single, trusted source of truth, with every metric defined once and defined consistently. In practical terms, it connects marketing spend to quotes, to policies, and to customer lifetime value, and it is what marketing leadership reads when deciding where campaign dollars go. Built entirely on Microsoft Fabric, this is a hands-on build-and-own role: the person in this seat decides how the semantic layer is structured, defends those decisions to the teams that depend on them, and is the escalation point when the model and the business disagree about what a number means. This is not a report building role and it is not a request queue.

Key Responsibilities

• Design, develop, and maintain the data pipelines and backend queries that populate the semantic layer, maintaining separation between raw, clean, and reporting layers so business logic lives where it belongs and not in the reporting layer.
• Own the data modeling and architecture of the semantic layer end to end: structure, relationships, storage mode, and refresh behavior, including the architectural decisions behind each.
• Author, maintain, and performance-tune the DAX measure layer, including the measure patterns downstream report builders depend on.
• Implement and maintain data frameworks and architectures that keep the platform's data consistent, accurate, and ready for use.
• Produce and maintain model documentation as a first-class deliverable, since the semantic layer serves report builders, analysts, and stakeholders who did not build it.
• Combine, optimize, and manage multiple upstream data sources, partnering with the business intelligence and data engineering teams on source changes, deployment practices, and promotion of work from development into production.
• Serve as the technical point of contact when downstream consumers report the model is wrong, and own the investigation through to root cause and fix.
• Perform on-demand analysis of complex data to identify strategic opportunities and efficiencies and to keep key business metrics accurate and trustworthy.
• Mentor apprentice team members working on model quality assurance and report migration, and review their work.
• Contribute to the department's applied AI efforts, including agent-based access to model documentation and semantic models.

Required Qualifications

• Strong SQL, including T-SQL, sufficient to write, review, and debug production analytical queries, and to recognize when generated or inherited code is incorrect rather than merely plausible.
• Python, sufficient to build and maintain data transformation notebooks.
• DAX, including measure authoring, performance tuning, and evaluation context.
• Demonstrated experience owning a large, centralized semantic model or semantic layer, or an equivalent analytical data product, including responsibility for its structure rather than only its contents.
• Experience working with data engineering and business intelligence partners on shared infrastructure.
• Ability to make and defend architectural decisions independently, and to explain the tradeoffs to both technical and business audiences.

Preferred Qualifications

• Microsoft Fabric experience (pipelines, lakehouses, warehouse, Direct Lake semantic models). Fabric experience is rare and learnable on the job by a strong candidate, so this is preferred rather than required.
• Tabular Editor.
• Marketing measurement or attribution background: how spend connects to quotes, binds, policies, and lifetime value.
• Insurance industry experience.
• Experience building or operating data quality frameworks.

Skills DAX, Data Analytics, Data Modeling, Data Pipelines, Data Quality, Large Semantic Models, Microsoft Fabric, Microsoft Power BI, Python, Semantic Data Modeling, Semantic Layer, SQL, Technical Documentation, Tabular Editor

Experience

• 3 or more years of experience (Preferred)

Supervisory Responsibilities

• This job does not have supervisory duties. It does carry technical mentorship responsibility for apprentice team members.

Skills
Big Data Analytics, Business Intelligence (BI), Data Analytics, Data Engineering, Data Modeling, Data Pipelines, Data Quality, Data Transformation, Python (Programming Language), Semantic Modeling, Structured Query Language (SQL)

Compensation
Compensation offered for this role is 70,100.00 - 121,475.00 annually and is based on experience and qualifications.

Allstate provides a comprehensive technology setup, including a laptop, monitors, headset, keyboard, and mouse. Employees eligible to work from home also receive a monthly connectivity reimbursement to help offset internet costs.

When working from home, you must have a dedicated, private workspace free from distractions, along with appropriate desk and seating. Reliable internet is required, with minimum speeds of 50 MB download and 5 MB upload.

To help create a more personal and engaging experience, we ask that you join with your camera on if your interview is virtual. Please also have a Valid Photo Identification available at the start of your interview. If you require any accommodations or have questions ahead of your interview please reach out to your recruiter.

Note: Internal candidates are only asked to join on video; a Valid Photo Identification is not needed for the interview.