Senior Analytics Engineer

Senior Analytics Engineer
Who we are

Jellyvision is redefining how organizations experience benefits by bringing everything together in one modern, intelligent home. With ALEX Home, we combine our award-winning ALEX® decision support with a flexible benefits administration platform, giving employers and employees a simpler, smarter way to manage benefits year-round.

Our mission is to help organizations reduce complexity, lighten administrative burden, and drive real employee understanding and utilization without forcing rip-and-replace decisions. We meet teams where they are today and give them a clear path to what’s next.

The people behind Jellyvision are creative problem solvers who care deeply about getting it right. We debate ideas, give real feedback, and sweat the details because those details are what turn complicated problems into great experiences for real humans.

We’re a human-first company that trusts smart people to do great work. We value curiosity, kindness, and willingness to try new things, learn fast, and try again. You won’t just show up to do a job, you’ll help build what’s next, solve real problems, and have some fun doing it.
What’s the role?

As a Senior Analytics Engineer, you'll own the layer between our data platform and the business — the models, metrics, and dashboards that turn raw data into something people trust and act on. Working closely with our platform data engineers, who own ingestion, pipelines, and infrastructure, you'll pick up where they leave off: shaping warehouse data into well-modeled, well-tested datasets in dbt, and building the dashboards and reporting in our BI layer that leadership, product, and operations rely on every day. The modeling is the craft; the dashboards are the point — you'll be responsible for both, and for making sure the numbers people see are ones they can trust.

This is a role for someone who loves the analytics engineering discipline and brings real engineering rigor to it — clear project structure, sensible conventions, and a considered point of view on what belongs in dbt versus the BI layer. You move fast and iterate frequently, and you're just as comfortable shaping something new as bringing order to what already exists.
What you’ll do to be successful
1. Build durable, trusted data models

- Design and build data models in dbt — staging, intermediate, and mart layers — with sound structure, incremental logic, tests, and documentation

- Apply dimensional modeling and grain discipline (star schemas, slowly-changing dimensions) so models are correct, performant, and reusable

- Define core business metrics once, correctly, and in a way the whole organization can rely on

- Classify sensitive data, maintain lineage and documentation, and define data-quality expectations at the model layer, partnering with the platform team on enforcement

Success looks like: Models are trusted, well-tested, and reused rather than rebuilt. Metric definitions are consistent everywhere they appear.
2. Own the semantic layer and metric definitions

- Build and maintain the semantic layer between marts and dashboards, so a given metric means the same thing across every report

- Partner with the business to define metrics and reconcile competing definitions into a single source of truth

- Keep modeling documentation clear enough that anyone can understand what a metric means and how it's derived

Success looks like: The business argues about decisions, not about which number is right .
3. Deliver BI the business trusts

- Build dashboards and reporting in our BI stack (Omni), grounded in well-modeled data rather than one-off queries

- Handle row-level security, performance, and access so dashboards are reliable and appropriate for a regulated, multi-tenant environment

- Translate ambiguous business questions into modeled metrics and clear, usable dashboards, and communicate data caveats plainly

- When a dashboard looks off or a number doesn't reconcile, yo