ML Engineer - Data Enrichment for Autonomous Driving

Wayve; Greater London; Hybrid; GBP 85,000 - 130,000; Full time

Benefits offered by this job

Hybrid working

Equity

Relocation support

Health insurance and wellbeing

Job summary

Wayve is building the leading AI platform for autonomous driving. This Data Enrichment role focuses on ground-truth datasets and classifiers describing road geometry, behavior, and scene attributes to support safe, scalable perception models.

You will collaborate across teams to define data requirements, curate high-quality datasets, and push improvements from experimentation to production, with a hybrid working model and opportunities to impact real-world driving systems.

Qualifications

• Proven experience developing and evaluating computer vision models for image or video understanding.
• Strong Python programming skills and PyTorch experience.
• Experience curating training and evaluation datasets and assessing annotation quality.

Responsibilities

• Define scene and scenario attributes with internal customers and establish measurable acceptance criteria for annotations.
• Design and refine annotation tasks and labeling instructions, ensuring quality through pilots, annotator agreement and expert reviews.
• Curate and deliver ground-truth datasets for training/evaluating models across countries and vehicle platforms.
• Build and evaluate computer vision models using Wayve's World Foundation Model to identify scene attributes and driving behaviours.
• Analyze errors to identify data, label or model gaps and test improvements.

Skills

Python

PyTorch

Computer vision

Dataset curation

Model evaluation

ML deployment

Job description

Wayve is building the leading AI platform for autonomous driving. This Data Enrichment role focuses on ground-truth datasets and classifiers describing road geometry, behavior, and scene attributes to support safe, scalable perception models.

You will collaborate across teams to define data requirements, curate high-quality datasets, and push improvements from experimentation to production, with a hybrid working model and opportunities to impact real-world driving systems.