Build and evaluate the assessment pipeline that turns editorial standards into evidence-backed findings.
About the role
This is a founding role. You will work directly with the product and editorial leads, with significant autonomy and significant responsibility for getting the details right.
What you’ll do
- Design and maintain the assessment pipeline and evaluation harness.
- Build calibration and drift monitoring for per-dimension scores.
- Work with editorial leads to translate standards into measurable criteria.
- Publish model cards and evaluation reports with limitations.
- Review code, write design documents, and mentor engineers.
What you bring
- 5+ years building production ML systems.
- Strong Python and experience with evaluation, not just training.
- Familiarity with NLP evaluation, calibration, and uncertainty.
- Ability to explain model behaviour to non-technical editors.
- Careful written communication.
Nice to have
- Experience in news, fact-checking, or content moderation.
- Familiarity with the EU AI Act or GDPR.
- Open-source contributions in evaluation tooling.
What we offer
Planned benefits at launch: health cover, pension contribution, a learning budget, a home-office allowance, parental leave for all parents, and dedicated time for public writing or open source. Policies are not final and will be published in full before any offer.
How to apply
Email careers@mediameter.eu with a short note and something that shows how you work. We read every application and reply to all of them. If you need an adjustment to the process, tell us and we will arrange it.
Equal opportunity employer
MediaMeter assesses candidates on the work and does not discriminate on the basis of age, disability, ethnicity, gender identity, nationality, religion, or sexual orientation.