Data Engineer II

We're Scan.com . We're building America's national imaging network: the infrastructure that connects patients, providers, and payers to faster, better-value diagnostic imaging. Thousands of patients move through our platform every day, and each is a real person waiting for an answer to a healthcare concern. We exist to get them that answer faster.

We manage the medical imaging pathway end-to-end. Providers order through a single point of entry, and from there, we run the entire medical diagnostic pathway. Managing the full pathway, rather than brokering pieces of it, is what lets us deliver imaging faster, at higher quality, and at a lower cost than the fragmented status quo.

We're looking for a Data Engineer at an exciting time. We've raised over $200m in funding to date, reached profitability, and are growing 100% year on year.

Millions of people wait too long for a diagnosis. Come help us fix that.

WHAT YOU WILL BE GETTING INVOLVED IN

Data at Scan.com is a full-stack discipline. We don't have analysts who hand off to engineers to build the model, or engineers who hand off to analysts to build the dashboard. As a Data Engineer , you will own the work end-to-end, from raw source data through the transformation layer, through the product that a stakeholder uses to make a decision.

Our data team supports every function across the US and UK: operations, revenue cycle, marketing, sales, provider success, and product. We sit at the center of a fast-moving business, and the work ranges from answering a precise operational question to building the automated reporting infrastructure that makes that question answerable without us in the loop next time.

We use a modern tech stack: Fivetran and Python for ingestion, Snowflake as our data warehouse, dbt for transformations, GitHub for CI/CD, Tableau for visualizations, and Cursor, Claude, and Snowflake Cortex for AI-assisted development. Our source systems span Postgres, HubSpot, Front, Acuity, Dialpad, and Facebook and Google Ads.

As a scale-up business, you can expect your role to develop over time. Here are some of the types of things you could be getting involved in:

Build and maintain dbt models that transform raw source data into reliable, well-documented, tested analytical assets used across the business

Partner with stakeholders across Operations, Finance, Marketing, Sales, and Product to translate ambiguous business questions into precise analytical deliverables. You should be able to identify if the ask needs refinement before the work begins

Design and ship Tableau dashboards and data products that automate reporting that currently requires manual effort, permanently removing recurring analytical burden from the team

Identify gaps in our data coverage and work with engineering or independently to close them. This can mean writing a Python ingestion pipeline or modeling a new source

Maintain data quality standards across the warehouse: write dbt tests, monitor for anomalies, and ensure that the numbers stakeholders see are trustworthy

Contribute to the team's analytical infrastructure. This means shared macros, source definitions, documentation standards, and CI/CD practices.

Stretch into various data science or data engineering projects based on business demand and based on your own personal development: pipeline development, ML feature preparation, or predictive modeling.

THE TOP 5 THINGS WE WANT YOU TO ACHIEVE IN YOUR FIRST YEAR ARE

Deep business context. Within 90 days, you understand the unit economics, operational workflows, and key performance drivers across the functions you support. You can receive a stakeholder request and immediately identify whether it's answerable with existing models, requires new modeling work, or requires a better-defined question.

Reliable analytical infrastructure. You have meaningfully improved the coverage, test depth, and documentation of our dbt model layer — making it easier for the next analyst to build on your work and reducing the frequency of data quality incidents.

High-impact data products shipped. Several Tableau dashboards or analytical products are in production and actively used by stakeholders to make decisions, replacing manual reporting or filling a genuine analytical gap.

Automation wins. You have identified and eliminated at least one recurring manual reporting process, replacing it with a scheduled, automated output that runs without human intervention.

Trusted analytical partner. Stakeholders across the business seek out your input before they finalize requirements, not after. You are known for asking the right clarifying questions and delivering work that answers the real question, not just the stated one.

WHAT YOU MIGHT BRING TO THE TABLE

You don't need to tick all the boxes to apply for this role. Whether it's your first role or your fifth, we believe everyone can add value, learn, and grow. However, these might be some of the ways you are currently adding value:

Strong SQL: you write complex queries fluently, understand query performance, and don't need a template to construct a multi-stage transformation

Hands-on experience with a modern data transformation tool, ideally dbt or SQLMesh. You understand the model DAG, have written tests and documentation, and are familiar with CI/CD practices in a transformation layer

Experience with a modern cloud data warehouse, ideally Snowflake or BigQuery — you understand how storage and compute interact and can write warehouse-idiomatic SQL

Proficiency with a BI tool such as Tableau, Metabase, Sigma, or Omni — you can design a dashboard that stakeholders actually use, not just one that technically answers the question

Strong stakeholder instincts: you have worked directly with non-technical stakeholders to scope and deliver analytical work, and you know how to translate between business language and data model logic

Python for data work. Ingestion pipelines with libraries like dlt, pandas-based transformation scripts, or scripted automation of manual reporting processes

For this role, the expectation is that you are an AI-Native. You build with AI directly, shipping, automating, or prototyping real work with LLMs and agents, and have clear judgment on where humans must stay in the loop.

Startup experience is a plus. You are comfortable with ambiguity, can prioritize without perfect information, and don't need a fully defined ticket to get started

Healthcare experience is a plus, particularly in imaging, RCM, or provider operations — but strong analytical fundamentals in any high-velocity domain are equally valued

AI REQUIREMENTS

At Scan.com , AI isn't a side project; it's how we work. We expect everyone to use AI to think faster, remove drudgery, and multiply their output. What we care about is judgment: knowing when AI makes you sharper, when it doesn't, and how to use it responsibly around patient data. We'll assess this in the way that matters for this role, using the tier below. You don't need to be an engineer. You need to be someone who reaches for the best tool and gets more done as a result.

Tier 1 — For this role, the expectation is that you are an AI-Native. You build with AI directly, shipping, automating, or prototyping real work with LLMs and agents, and have clear judgment on where humans must stay in the loop. (this is typically aligned to engineering, data, product, ops automation, RevOps)

You build with AI, not just alongside it. You've shipped, automated, or prototyped something real using LLMs, agents, or AI tooling.

You can explain a workflow you redesigned so a machine does the boring 80%.

You have a view on where agentic systems help and where a human must stay in the loop, especially in regulated, PHI-adjacent work.

Assessment: a practical exercise where you use AI tools live or in a take-home.

A VIEW ON OUR CULTURE

We're building something that matters. Every scan we speed up is a real