Data Scientist, Data Quality & Provenance Team at Wayve
Germany
Posted
on-site · full_time · Visa sponsorship
Responsibilities:
- Define measurable, statistically rigorous data-quality concepts (coverage, label quality, uncertainty, provenance, performance).
- Develop and apply statistical inference methods for multi-rater annotation, label ambiguity, and noisy evaluation signals.
- Build automated reports and reusable metrics that communicate confidence, limitations, and decision implications to stakeholders.
- Analyse large annotation and dataset pipelines using Python and SQL to surface quality issues and drive remediation.
- Work with annotation, autonomy, and evaluation teams to integrate metrics into platform workflows and tooling.
Requirements:
- Strong foundations in classical statistics, experimental design, sampling, estimation, and inference.
- Practical experience with annotator/rater modelling, inter-rater agreement, label uncertainty, or related methods.
- Proficiency in Python and SQL and experience analysing large, complex datasets.
- Familiarity with probabilistic/Bayesian methods and human-in-the-loop ML systems preferred.
- Ability to translate ambiguous stakeholder questions into defensible metrics and clearly communicate uncertainty and trade-offs.
- Experience with annotation tooling, QA methodologies, or simulation-based evaluation is a plus.