Analytics Engineer

About Northwind Analytics

Northwind Analytics builds a warehouse-native analytics layer so that finance and operations teams can answer their own questions instead of filing a ticket and waiting a fortnight. We are about 140 people across twelve time zones. Almost nothing here is decided in a meeting; proposals circulate as documents, get argued over in comments, and then someone builds the thing.

This role sits on the internal data team, which means you are both a user of our product and one of its most demanding critics.

What you'll do

• Own the modelling layer for product usage, billing and customer health data, currently around 400 dbt models with a maintenance debt we would like you to reduce.
• Redesign our event schema alongside the product engineering team so that instrumentation is decided before a feature ships rather than bolted on afterwards.
• Build the metric definitions that go on to appear in board reporting, and defend them when the numbers are inconvenient.
• Set up testing and observability for our pipelines so we hear about a broken model from a monitor rather than from a confused account manager.
• Partner with go-to-market teams on retention and expansion analysis, including the unglamorous work of agreeing what counts as an active account.
• Feed real problems back into our own product roadmap, since we are an obvious test case for everything we ship.

What we're looking for

• Four or more years in analytics engineering, data engineering or a heavily technical analyst role.
• Excellent SQL. Not just correct queries, but queries that stay understandable six months later and do not cost a fortune to run.
• Production experience with dbt or an equivalent transformation framework, including how you structure a project that several people contribute to.
• Comfort with Python for the parts a warehouse cannot do well.
• The instinct to ask what a stakeholder is actually trying to decide before writing any code.
• Clear written communication and the discipline to document a model while you still remember why it exists.

Nice to have

• Experience with Snowflake or BigQuery at a scale where cost management mattered.
• Familiarity with subscription revenue concepts such as recognised revenue, churn cohorts and net retention.
• Exposure to a semantic layer or metrics store.

Benefits

• Health insurance for you and your dependents, or a regional cash equivalent.
• Retirement contributions with company matching where available.
• Sixteen weeks of paid parental leave for all new parents, regardless of how the child joins the family.
• A genuinely async culture, with two optional overlap hours a day and everything else written down.