← back to main site

gustav sparre

data engineer

gothenburg mail github linkedin

summary

problem solver who ended up in data engineering because pipelines are just puzzles with consequences. naturally curious, chronically tinkering – the type to rebuild something three times until it feels right. takes full ownership of what i build because half-done solutions are worse than no solution. drawn to data work because the feedback is honest: it works or it doesn't, no arguments.

treat code like craftsmanship – every pipeline should reflect care and intent. want to work with people who share that ambition, who see their work as more than just tickets to close. focused on enabling both stakeholders and peers to excel, because good solutions matter more when everyone can build on them.

technical skills

orchestration
Dagster, Airflow
data processing
Apache Spark, dbt, SQL, Python, Polars, duckDB
data platforms
Snowflake, Iceberg
infrastructure
Kubernetes, Helm, Docker, CI/CD, GitHub Workflows
languages
Python, SQL, Bash
other
Git, Nix, Data Modeling

experience

2025 – present

data engineer

volvo cars

  • killed powerbi, replaced it with evidence. stakeholders can actually find their data now
  • migrating to iceberg because our 'data lake' was actually a swamp
  • rebuilt ci/cd from failed git-flow hell to working trunk-based deployment
  • cleaning up technical debt and ensuring scalability for future growth
  • working with stakeholders to improve data quality and accessibility (data products)
  • mentoring peers and promoting best practices in data engineering
  • looking into new tech and tools to keep the platform modern and efficient
2023 – 2025

graduate / data engineer

knowit solutions cocreate

  • worked with pipelines processing sweden's entire personal data registry (dun & bradstreet)
  • built data pipelines and analytics for the software being installed in cars (volvo cars)
  • redesigned the graduate program twice because someone had to

education

2015 – 2019

bsc, mechanical engineering

chalmers university of technology

the formulas are long gone, but the concepts stuck. what actually mattered wasn't memorizing how to calculate entropy or sizing bolts – it was learning how to break down problems logically and teach myself whatever i needed to know. turns out that's way more useful than any specific equation.

focus: mathematics, physics and programming