Senior Data Engineer - Analytics

Machinify is hiring a Senior Data Engineer (Analytics) to turn healthcare operational data into trusted, governed models, metrics, and reports. The role partners with business stakeholders and owns data modeling, metric definitions, visualization, observability, orchestration, documentation, and automation.

Responsibilities:

• Run discovery with business stakeholders. Find the real need behind the request and turn it into acceptance criteria you can build against
• Model noisy healthcare operational data into certified dimensional models across our Databricks bronze → silver → gold architecture in dbt and SQL. Our sources disagree with each other in ways that are interesting rather than trivial. Sorting that out is most of the job
• Own metric definitions. One measure, one definition, one place, documented so a non-engineer can read it. Today a single operating metric can be re-derived in a dozen models; collapsing that into a governed semantic layer is yours
• Build and own the visualization layer. Power BI semantic models, dashboards, and paginated reports, including refresh health and semantic-model currency. We want real visualization judgment: the right chart for the question, a hierarchy an executive can read in ten seconds, and a reason for every choice
• Own observability for your models. Freshness and latency detection, statistical anomaly detection, and value-drift monitoring, plus the data tests, regression guards, and CI checks that keep a defect you just fixed from coming back
• Own orchestration for what you build. dbt Cloud job design, Airflow DAGs, scheduling, dependency ordering, dependable incremental runs
• Write it down. Design documents, model documentation, a CI-enforced knowledge base, and business-readable handoff packets, so whoever inherits your work doesn’t have to re-derive it
• Automate the recurring work. We codify repeating work into repeatable, testable, AI-assisted workflows instead of checklists in someone’s notes. You should be comfortable directing LLMs and supervising what they produce

Requirements:

• 6+ years across analytics engineering, BI engineering, data engineering, or business analysis, with production ownership of the models and metrics you built
• Expert SQL and strong dimensional modeling instincts
• Production dbt: models, tests, macros, exposures, documentation, and CI
• Hands-on with a cloud lakehouse or warehouse
• You've owned a BI semantic layer and the reports on top of it, not just the tables underneath
• You can run a requirements conversation with a non-technical owner, turn a vague complaint into a testable definition, explain a data problem as business risk, and tell someone their metric is wrong without losing them
• A reconciliation reflex. When two numbers disagree you don't average them, escalate them, or trust the newer one. You find out why
• Git and pull-request discipline as normal practice rather than overhead
• Python where it's genuinely the right tool (API calls, file parsing, statistics, report generation), and the judgment to reach for SQL where it isn't
• Comfort with high autonomy in a fast, async, Slack-first environment. You decide inside your lane and inform, rather than queuing decisions for permission
• Strong writing. Much of your impact is a document, a definition, or a chart someone reads without you in the room
• Databricks and Spark SQL preferred; Snowflake, BigQuery, or Redshift transfer well
• Power BI and DAX strongly preferred; Looker/LookML, Tableau, MetricFlow, or Cube are credible substitutes
• Healthcare data: claims (837/835/UB04), DRG and coding review, payment integrity, provider contracts, or revenue cycle
• Data contracts, schema enforcement, or ODCS-style governance; quality gating in CI
• Databricks Unity Catalog, lineage, and cost/DBU accountability
• Paginated reporting, SharePoint/Excel delivery, or other “the business needs it in exactly this format” surfaces
• Working alongside AI coding agents and reviewing what they produce against a contract you wrote