Data Engineer

Skatteguiden; København; On-site; DKK 700,000 - 900,000; Full time

Benefits offered by this job

Real ownership

Modern stack

Office in Copenhagen

Job summary

Skatteguiden is seeking a Data Engineer to own end-to-end data ingestion pipelines and governance of the data platform. You’ll work closely with product and platform teams to shape data contracts and keep pipelines resilient as systems evolve.

You will own Dagster orchestration and Snowflake infrastructure as code, with responsibilities ranging from environments and access control to cost management. The role offers a modern stack and direct impact on Danish tax insights.

Qualifications

• Experience owning end-to-end data ingestion pipelines, including handling schema drift and backfills.
• Experience owning orchestration in a tool like Dagster, Airflow, or Prefect and owning failure recovery.
• Experience managing cloud data infrastructure as code—Terraform required; DCM preferred.

Responsibilities

• Own end-to-end data ingestion pipelines from source systems into Snowflake.
• Own orchestration in Dagster, including scheduling, dependencies, observability, and failure recovery.

Skills

Data ingestion pipelines

Dagster orchestration

Cloud data infrastructure as code

dbt and data warehouse

Cross-team collaboration

Tools

Dagster

Airflow

Prefect

Terraform

DCM

dbt Cloud

Snowflake

Job description

Why this role exists

Tech-first. Ego-free. User-obsessed. Over a million Danes use Skatteguiden to understand and optimise their taxes — automating tax tracking, surfacing deductions people didn't know they had, and connecting directly to the Danish tax authority to do it. We’re an independent, 50-person company based in Copenhagen with a 4.7-star rating and one mission: make financial insight accessible to everyone, not just those who can afford an accountant.

None of that works without a data platform people can trust. Ours is built on Snowflake, Dagster, and dbt, turning messy source data into the numbers people rely on. We’re investing further in it, and we need a data engineer who can take real ownership of it — from infrastructure and ingestion down to the SQL.

What you’ll be working on

• Own end-to-end data ingestion pipelines — from source systems into Snowflake — partnering with our product and platform teams to shape data contracts and keep pipelines resilient as those systems evolve.
• Own orchestration in Dagster — scheduling, dependencies, observability, and failure recovery across the pipeline.
• Own our Snowflake infrastructure and custom components as Infrastructure-as-Code — provisioning, environments, access, cost — using Terraform (and ideally DCM).
• Push our data engineering practices forward — testing, CI/CD, semantic layer modeling, performance on Snowflake.
• Shape how we use AI in the data stack.
• Design and maintain dbt models that turn raw Snowflake data into trustworthy, well-tested data products — including owning dbt Cloud itself: environments, jobs, and deployments, not just the modeling layer.
• Get exposure to Java (or another OOP language) and event-driven architecture.

What you bring

• A track record of owning data ingestion pipelines end-to-end, including the messy parts — source system quirks, backfills, schema drift.
• Experience owning orchestration in a tool like Dagster (or similar — Airflow, Prefect), and comfort being the one who gets paged when a DAG fails.
• Experience managing cloud data infrastructure as code — Terraform required, DCM preferred.
• Strong hands-on experience with dbt and a cloud data warehouse — including dbt Cloud administration (environments, job configuration, deployment), not just model development.
• Comfort working directly with engineering teams outside of data, translating platform and product needs into data models.

Nice to have

• Java or another OOP language, and comfort with event-driven architecture.
• Exposure to applying AI/LLMs within a data platform.

What we offer

• Real ownership — a small team where what you design ships, infrastructure included, not one buried in a 40-person backlog.
• A modern stack, not legacy debt: Snowflake, Dagster, dbt Cloud, Terraform, Evidence, Lightdash — built recently, built to be extended.
• Work that sits close to product and platform, not isolated in a data silo — your input shapes what gets built upstream, not just what you transform downstream.
• A problem worth solving: Danish tax law is genuinely complex, and modeling it well is a real engineering challenge.
• Impact you can point to — your pipelines and models power a product over a million Danes actually use to get their taxes right.
• Office first in Copenhagen — five minutes from Rådhuspladsen, lunch at Claus Meyer, dinner covered if the day runs long.

Practical

Start: As soon as possible Reports to: Head of Data & Operations Planning

Hiring process

• 30-minute online screening call with recruiter.
• 1-hour onsite case session with hiring manager and team member.
• 30-minute onsite meet-and-greet with CEO.