Business Analytics Engineer, Customer Insights

About the role Rad AI is building out embedded analytics inside our reporting platform - giving radiology groups and health systems self-service visibility into their own reporting data: turnaround times, template usage, AI utilization, auto-launch accuracy, and more. This role sits at the center of that effort: part data engineer, part analyst, part customer-facing partner. You'll work with customers to understand what they actually need to see, then build the governed data models and dashboards that deliver it correctly, at scale, across customers with very different data volumes, specialties, and reporting patterns. This is not a back-office reporting job. You'll work with customer stakeholders (radiology administrators, quality leads, IT/informatics teams), with Product and an embedded analytics partner alongside you, translating ambiguous asks ("we want to see if our new templates are actually helping") into rigorous, reusable metrics and explaining them when a customer's data looks stranger than expected. What you'll do • Get the data foundation right. Connect the new data platform to the existing warehouse, troubleshoot and fix data quality issues per customer, and help define the data contract that doesn't exist yet. • Gather and translate requirements with customers and internal partners. Join discovery conversations with radiology group administrators, quality/ops leads, and informatics teams to understand what they're trying to measure and decide, not just what they initially ask for. • Design and build the semantic layer and dashboards that serve customer requirements: governed metrics, joins, and reusable models that hold up across customers rather than one-off reports per account. • Apply real statistical rigor to healthcare reporting data. Radiology data is messy and high-stakes: small sample sizes at some sites, outliers from edge-case studies, seasonality, case-mix differences across specialties and facility types. You'll need to know when a metric needs a confidence interval, a risk-adjustment, or an exclusion note, not just a number. • Use AI as a core part of your workflow. Direct AI agents to describe, model, and document data relationships, and verify what they produce. • Own data quality end-to-end for customer-facing outputs: catch and explain anomalies before a customer does, document known caveats (sample-size exclusions, data lag, definitional edge cases) directly in the deliverable. • Build durable data products, not one-off answers. Use dbt/Snowflake-style modeling practices so a metric built for one customer's dashboard is reusable for the next, instead of re-derived from scratch each time. • Partner cross-functionally with Product, Engineering, and Customer Success to prioritize which metrics and dashboards actually move the needle for customers versus which are just "nice to have." • Explain data to non-data audiences. You'll present findings to radiologists, hospital administrators, and executives who are expert clinicians but not data scientists. The ability to make a confidence interval or caveat meaningful to that audience is as important as computing it correctly. What you'll bring • 2-5 years in an analytics engineering, data science, or business/BI analyst role, ideally with some customer- or stakeholder-facing exposure (not purely internal reporting). Hunger and curiosity matter more to us than years. • Demonstrated AI fluency. Concrete examples from the last six months of how AI tools have meaningfully changed the way you work with data, including how you've used them to describe, model, or document datasets and how you check their output. • Strong SQL - comfortable writing and optimizing complex queries, not just filtering existing tables. • Working statistics fluency: confidence intervals, significance testing, sample-size considerations, handling outliers and skewed distributions, and the judgment to know when a metric needs a caveat versus when it's solid. • Experience with a modern governed BI/semantic layer tool (we use Omni; Looker, Sigma, ThoughtSpot, or similar is great) and a SQL/Python notebook or exploration environment (we use Hex; Mode or similar is fine). • Comfortable in dbt/Snowflake-style modeling environments: building something once and having it hold up for the next ten customers. • Direct experience defining or maintaining a semantic layer (entity/metric definitions, joins, governed fields, e.g., LookML, dbt Semantic Layer, or an equivalent governed-BI modeling layer), not just consuming one someone else built. • Comfort with ambiguity. You're energized, not rattled, by undefined problems, messy data, and no existing playbook. • Demonstrated ability to run discovery with external stakeholders: asking good follow-up questions, pushing back on vague asks, and landing on a defensible metric definition. • Excellent written and verbal communication, able to present data findings to a non-technical, clinical audience without dumbing down the substance. • Healthcare, health IT, or other regulated-industry data experience is a strong plus (HIPAA awareness, understanding why "it depends on the payer/facility/specialty" is often the right answer). • Familiarity with radiology-specific concepts (PACS, HL7/FHIR, turnaround time, RVU, template-driven reporting) is a bonus, not a requirement. We'll teach the domain if the analytical and statistical instincts are there. Nice to have • Experience building or maintaining a customer-facing (vs. internal-only) analytics product, i.e., you understand that a dashboard a paying customer sees is held to a different bar than an internal Slack chart. • Exposure to multi-tenant data governance (row-level security, per-customer data isolation) in an embedded analytics context. • Prior experience in a company selling into hospitals/health systems, understanding procurement and IT review cycles. Location Details: For roles listed as San Francisco - Onsite: • This role will be based in our San Francisco office and we expect employees to work onsite four days per week. The remaining time may be worked remotely or onsite, depending on team and business needs. For roles listed as United States - Remote: • This role is open to candidates located anywhere in the United States. For roles listed as San Francisco - Onsite + United States - Remote: • We will prioritize candidates who can work onsite four days per week in San Francisco, while also considering remote candidates located anywhere in the United States. For US-Based Full-Time Roles, Rad AI offers a variety of benefits, including: • Comprehensive Medical, Dental, Vision & Life insurance • HSA (with employer match), FSA, & DCFSA • 401(k) • 11 Paid Company Holidays • Flexible PTO policy • Annual company-wide offsite • Periodic team offsites • Annual equipment stipend • For roles based outside the US, your recruiter can share more details