Data & Production Engineer
Today the models run as notebooks, scheduled jobs and a Worker. That holds until the first client licenses a live feed, and then somebody owns uptime, vendor ingestion, versioning, and the awkward email about why yesterday’s number moved.
Indicative range, set by experience and by where you are. We would rather say a number up front than waste your time.
What you would do
- Turn research code into monitored, versioned feeds a client can build on.
- Own ingestion from public and vendor sources, and the pipeline that keeps it fresh.
- Build the monitoring that tells us a number moved before the client tells us.
- Keep the open-source side something we are happy to have people read.
What you would bring
- Python in production, orchestration, time-series storage, CI you would trust at 6am.
- Experience shipping data products with a number attached, not just notebooks.
- Cloud infrastructure you can run alone. We are small and there is no platform team.
- Care about reproducibility, because the published research depends on it.
How we work
We are three people and we stay small on purpose. Everyone here builds, validates and ships their own work, and everyone talks to the people who use it. The trade-off is total ownership and very little structure. There is no manager, no HR department and no job ladder waiting for you.
How to apply
Send a short note and your CV to careers@causalsystems.co. Tell us what you have built and what you would want to build here, and attach anything worth reading: a paper, a repository, a backtest, a deck. No cover letter theatre and no form to fill in.
We read everything and we reply, including when the answer is no.