Director, data platform

Ph3Job Type /h3 pFull‑time /p h3Description /h3 pAvonRisk is a national provider of risk management, claims administration, and managed care services for carriers, employers, and public entities across the United States. We unite respected regional leaders in workers' compensation, liability, managed care, and risk management — including Intercare, InterMed, George Hills, and ASG Claims Administration — into a single national platform serving 32 states and nearly 700 professionals. /p pWe’re not a "big box" TPA. We invest in tools and workflows that reduce friction — not increase volume — and create real career paths for people who want to build, not babysit. We’re a Microsoft‑first, AI‑enabled organization, and we’re building the modern data foundation that turns operational data into decisions our clients trust. /p h3Position Summary /h3 pThe Director of Data Platform owns AvonRisk’s enterprise data platform — the Azure Fabric lakehouse, semantic models, governance framework, and customer‑facing analytics that power every decision in our business. This is the most important technical hire in our 2026 transformation plan. You’ll architect and build the platform during a focused 6‑month surge with a contracted delivery team, then run and grow it as a permanent capability. /p pThis is a senior individual‑contributor leadership role, not a people‑management role. Success is defined by platform impact, architectural quality, and business outcomes — not headcount. You will be hands‑on in Fabric, Power BI, schemas, and security from day one. /p pThe 6‑month build is the foundation. The real impact comes from what the platform enables next — transformational analytics, predictive models, and client‑facing intelligence that create competitive advantage for our clients and operating leverage for AvonRisk. /p h3What You’ll Do /h3 h3What Success Looks Like in the First 90 Days /h3 ul liA clear, defensible Fabric architecture with environments, security, and semantic strategy defined /li liA surge team that is unblocked, productive, and shipping on a predictable cadence /li liAt least one real client dataset flowing end‑to‑end into a governed semantic model /li liEarly trust from BI, product, and operations that this platform will make their work easier, not harder /li /ul h3Build the Platform (Months 1–6) /h3 ul liOwn the Azure Fabric architecture end‑to‑end: workspace topology, OneLake / Iceberg storage strategy, dev/test/prod separation, capacity sizing, governance. /li liLead a 4‑person surge team (Data Architect, two Data Engineers, Business Analyst) through a 6‑month build. Set the bar; remove blockers; make them productive immediately. /li liDefine the Bronze ? Silver ? Semantic layered architecture, with a Gold semantic schema expressed in AvonRisk business terms (not source‑system terms) so reports, dashboards, and AI models survive future claims‑platform migrations. /li liDrive the Fabric Mirroring validation for our claims system (CE/iVOS) and design the ADF fallback where Mirroring isn’t viable. /li liArchitect multi‑tenant PHI security: client‑segregated PHI workspace + masked cross‑tenant analytics layer, with RLS and Fabric secure data sharing for client portal access. /li liStand up the customer‑facing analytics portal (Power BI Embedded or Fabric‑native) with a 10+ client pilot live by end of Month 6. /li liEstablish data catalog, lineage, DQ, and governance using Microsoft Purview and Fabric‑native tools. /li /ul h3Run and Grow the Platform (Month 7+) /h3 ul liOperate the platform as a permanent capability — uptime, performance, capacity, cost. /li liOnboard additional source systems as AvonRisk grows through acquisition and new programs. /li liBuild the AI/ML feature foundation that underpins our agent roadmap — Fraud, Provider Intelligence, Recovery, Safety Loss Control. /li liServe as the strategic technical voice for the transformational analytics built on top of this platform — cross‑client benchmarking, predictive severity, provider performance, client risk recommendations, litigation propensity, subrogation identification at scale. You won’t deliver every one of these yourself; you’ll be the thought leader on what’s possible, what’s worth doing, and what the business and our clients should expect from data we own. /li liPartner closely with product builders, claims operations, and client‑facing teams to keep the semantic model honest about what the business actually needs. /li /ul h3Enable the BI Team /h3 ul liThe BI team is a separate organization focused on business reporting today. Your job is not to absorb them — it’s to give them a platform that makes their work easier and unlocks their next chapter: /li liIn the near term, lift the BI team off hand‑built SQL and SSRS by giving them clean, governed semantic models they can self‑serve against. /li liOver time, partner with BI leadership as the team graduates from operational reporting into the higher‑value, more technical work the platform unlocks: data insights, advanced analytics, AI/ML on top of the lakehouse, and data science. Your platform is the foundation that makes that future possible. /li liCoach on dimensional modeling, semantic design, and Power BI best practices where it helps — without owning their team or their roadmap. /li /ul h3Lead Without a Hierarchy /h3 ul liWrite the documentation and runbooks that let this platform survive without you. /li liBe the technical voice in roadmap conversations with the CPTO, BI leadership, and operations leaders. /li liSet the standard for how data work gets done at AvonRisk — governed, reproducible, AI‑assisted, and built to last. /li /ul h3Requirements /h3 h3What You Bring: /h3 h3Required /h3 ul li7+ years of hands‑on data platform work, including 3+ years on a modern cloud lakehouse or warehouse (Fabric, Snowflake, Databricks, BigQuery, or Synapse). /li liProduction Azure Fabric experience preferred — alternatively, deep Azure data experience (Synapse, ADF, ADLS Gen 2, Power BI Premium) and a credible plan to come up to speed on Fabric in week one. /li liPower BI semantic modeling at depth — DAX, dataset design, Direct Lake mode, Row‑Level Security. You can architect a semantic model, not just consume one. /li liModern data engineering posture: DBT (or equivalent — SQLMesh, Coalesce) for transformation engineering /li liData testing frameworks (dbt tests, Great Expectations) /li liADF, Fabric Pipelines, or equivalent orchestration /li liCI/CD for data /li liInfrastructure as code (Bicep / Terraform) /li liGit‑based workflows for analytics code. We are explicitly not hiring "I know SQL and have been around data for 25 years" — we’re hiring modern analytics engineering. /li liDimensional modeling judgment — when to denormalize, when to conform, when to wait. Star schemas and slowly‑changing dimensions in your sleep. /li liMulti‑tenant + PHI/PII data architecture experience. RLS, masking, lineage, audit. You’ve done this for real, not just read about it. /li liStrategic data product thinking. You can articulate what business problems data uniquely solves — and which it doesn’t. You can be a thought partner on AvonRisk’s analytics roadmap, not just the platform underneath it. /li liTrack record leading or quarterbacking contracted delivery teams — including offshore — through fixed‑window engagements. /li liExcellent written communication. You can produce a design document a CPTO or auditor can read. /li /ul h3Strongly Preferred /h3 ul liInsurance, healthcare, financial‑services, or regulated‑data analytics product background — HIPAA, SOC 2, state regulatory reporting; experience shipping analytics products that customers or regulators actually consumed. /li liExperience migrating off legacy SSRS / stored‑proc reporting estates onto modern semantic models. /li liSpark / PySpark notebook experience for non‑SQL transforms. /li liData contracts and schema‑enforcement experience between data producers and consumers. /li liHands‑on with Microsoft Purview for catalog and lineage. /li liFamiliarity with AI/ML feature stores and operationalizing predictive models on top of a lakehouse. /li liComfort with Claude Code or equivalent AI‑assisted development workflows — this is how AvonRisk’s technology team works. We expect you to amplify yourself with AI, not resist it. /li /ul h3Education /h3 ul liBachelor’s degree in Computer Science, Information Systems, Engineering, or equivalent practical experience. /li liMaster’s a plus, not required. /li /ul h3Why This Role Matters /h3 pToday, 26% of our engineering capacity is spent on hand‑built reports. 194 clients rely on one‑off SQL. Entire lines of business lack meaningful analytics. Our AI roadmap — Fraud, Provider Intelligence, Recovery, Safety Loss Control — is blocked until we have a unified data foundation. The first dashboard a client sees in our new portal will be built on the platform you architect. /p pYou’ll have executive air cover, a green‑lit plan, a funded 6‑month surge team, and direct access to the CPTO. Build it right, and it’s the foundation under everything AvonRisk does for the next decade. /p h3Job Type Location: /h3 pFull‑time, remote (US). Eastern or Central time zone preferred for stakeholder overlap. Occasional travel to AvonRisk offices (Glendale CA, Rocklin CA, Texas, and others) — expect ~4–6 trips per year. /p h3Salary Range /h3 p$190k-$220k + bonus /p pThe salary range listed is an estimate. Actual compensation will be determined based on several factors such as a candidate’s experience, qualifications, skill set, and work location. /p h3Benefits /h3 ul liComprehensive medical, dental, and vision benefits /li liCompany contributions to HSA and FSA plans /li liEmployer paid life and disability insurance /li li401(k) with company match /li liPaid time off (PTO) and company paid holidays /li liLearning and development opportunities that support real career advancement /li liEmployee assistance resources and a supportive culture that values balance and wellbeing /li /ul h3Equal Opportunity /h3 pAvonRisk is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, age, disability, veteran status, sexual orientation, gender identity, or any other status protected by law. /p pPursuant to the Los Angeles and San Francisco Fair Chance Ordinance, we will consider for employment qualified applicants with arrest or conviction records. /p /p #J-18808-Ljbffr