arcadia - Lead Analytics Engineer - Data Modeling & Quality
Upload My Resume
Drop here or click to browse · Tap to choose · PDF, DOCX, DOC, RTF, TXT
Requirements
• Bachelor's or Master's degree in Computer Science, Statistics, Business, Economics, or a related field • Advanced SQL: window functions, complex CTEs, aggregation patterns, performance tuning on columnar databases • DBT: hands-on experience authoring models, tests, macros, and yml documentation; familiarity with incremental strategies • Healthcare data literacy: working knowledge of claims data (professional, institutional, pharmacy), clinical data (EHR entities), and common quality dimensions (member months, coverage rates, null patterns) • Data quality mindset: ability to differentiate source data issues from transform issues, design systematic validation checks, and communicate data quality findings clearly • Clear communicator — able to translate technical findings for clients and non-technical stakeholders • Strong analytical judgment — you can look at a distribution and know when something is wrong • Ability to manage several projects simultaneously, leveraging AI tooling to stay organized and efficient • Genuine desire to learn and apply AI tools for operational efficiency • ## Would Love For You To Have • Experience with Spark SQL and Hudi table format • Familiarity with data quality monitoring tools • Comfortable operating in an AI-first environment using Claude to build/verify various day-to-day workflows • Exposure to population health analytics concepts: HEDIS measures, risk adjustment, value-based care metrics • Python scripting for data investigation and automation • Experience with Argo Workflows or similar orchestration platforms • Healthcare data standards: ICD-10, CPT, NDC, LOINC, NPI • ## What You'll Get • Work alongside a talented team on some of the most complex and rewarding challenges in healthcare data • Flexible, fully remote work environment with the resources and support to do your best work • Exposure to senior leaders • Be on the front lines of AI adoption — use cutting-edge tools to accelerate your work and shape how the team operates in an AI-first environment • Make a meaningful impact on healthcare data operations by improving the quality, reliability, and trustworthiness of data that drives patient care decisions • Be a part of a mission driven company that is transforming the healthcare industry • Become a member of the talented, energized, diverse and purpose-driven Arcadian Community
Responsibilities
• Independently triage and resolve pipeline data quality issues • Author at least one new DBT model or refactor an existing one to meet current modeling standards • Design a DBT test suite for a set of models lacking coverage • Understand the end-to-end pipeline from ingress through silver and gold, and be able to trace a data quality issue to its root layer • Arcadia is dedicated to happier, healthier days for all. We believe that there is a better healthcare world – one powered by data. Our platform transforms complex, diverse data into a unified foundation for health, helping organizations deliver better care, boost revenue, and lower costs. • We’re a team of fiercely driven individuals committed to making healthcare more sustainable—and we’re looking for passionate people to help us get there. • For more information, visit arcadia.io. • DATA MODELING & DBT DEVELOPMENT • Author, review, and maintain DBT models using Spark/Hudi from ingest through bronze and silver • Help clients understand their data model, assumptions, and limitations through intentional validation • Troubleshoot and fix issues, then write DBT tests to catch issues proactively • Optimize SQL performance for slow-running jobs • Partner with Data Engineering on Hudi table design, partition strategy, and incremental patterns • DATA QUALITY OWNERSHIP • Triage and classify data quality alerts, distinguishing source-level issues from transform-layer failures • Design and maintain volume monitors and DQ monitors (null rate, distribution, future-date checks) • Author and apply clinical DQ rules (entity volume, field coverage, LOINC coverage, referential integrity) and claims validation rules across silver and gold layers • Conduct quality reviews for connector promotions — evaluating silver entity coverage, validation rule pass rates, and bronze-to-silver transformation correctness • Own the ticket queue for DQ, attribution, hierarchy, and customer-specific data quality issues, writing clear customer-facing findings • CROSS-FUNCTIONAL QUALITY COLLABORATION • Lead data quality reviews during connector installation and promotion (UAT → PRD), including claims validation playbooks and null analysis • Partner with Data Engineering on root-cause triage for errors, ingress anomalies, and silver table issues surfaced through data quality monitoring • Coordinate with the Measure Implementation Team (MIT) when data quality issues affect quality measure scores • Contribute to and enforce data modeling standards across teams • Data modeling: DBT-Spark, SQL, Claude • Warehousing: Amazon Redshift, Apache Hudi, AWS Athena • Data quality: volume/DQ monitors, DBT tests • Orchestration: Argo Workflows, Airflow • Source control: Git / GitHub, PR-based review workflows • Observability: Grafana, Loki, Jira • Healthcare data: Claims (plan/professional/pharmacy), EHR (clinical entities), MPI
Benefits
• Arcadia's data platform powers population health analytics for health plans, ACOs, and provider groups across the country. As a Lead Analytics Engineer — Data Modeling & Quality, you sit at the intersection of data quality ownership and analytical data modeling. You'll own the SQL and DBT layer that transforms raw clinical and claims data into trusted, production-grade datasets, while also serving as the quality authority for the data those models produce. • This is a hybrid role — deeper SQL and DBT expertise than a traditional Data Health Professional, with a more analytical and model-focused scope than a Data Engineering role. You're less focused on pipeline infrastructure and more on the logic, shape, and trustworthiness of the data itself. • What Success Looks Like
No credit card. Takes 10 seconds.