SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Wayve is building the AI platform for autonomous driving, pioneering an end-to-end AI approach that enables vehicles to learn from real-world experience. The Enrichment and Curation team transforms fleet-scale driving data into high-quality training datasets, enrichments, and evaluation benchmarks for Wayve's AI Driver and foundation models.
In this role, you will build scalable methods to discover rare, high-value, and safety-critical scenarios in fleet data using embeddings, active learning, similarity search, model-assisted mining, and smart sampling to improve dataset coverage. You'll design pipelines for automated enrichment, labelling, deduplication, and data-quality monitoring. You'll run experiments to understand which data mixtures and curation strategies improve model performance and collaborate with foundation-model, evaluation, simulation, and infrastructure teams to move ideas into production. You'll also build benchmarks and slice analyses that expose model blind spots and guide the next data and modelling iteration.
You'll work with petabyte-scale real-world driving and simulation datasets, data engines and retrieval systems using Python, SQL, Spark, embeddings, and multimodal models, training-data strategy for embodied vision-language and driving models, automated enrichment and active learning, and evaluation datasets that connect data decisions to real-world driving outcomes.
Wayve operates a hybrid working model combining in-person collaboration with focused remote work. The company offers market-benchmarked salaries, meaningful equity, relocation support and visa sponsorship where applicable, hybrid working with access to vehicle workshops and labs, learning and development support, and comprehensive location-dependent health, family, retirement, and wellbeing benefits.
REQUIREMENTS:
- Strong machine-learning and software-engineering fundamentals with ability to take applied research into production
- Proficiency in Python and a modern deep-learning framework
- Comfortable with SQL or large-scale data tools
- Experience with data curation, foundation models, large-scale data wrangling, computer vision, or model evaluation
- Ability to reason about sampling, class imbalance, label quality, dataset coverage, and experimental design
- Enjoy ambiguous, cross-functional problems and communicate technical trade-offs clearly
- Valuable but not required: experience with autonomous driving, robotics, VLMs, active learning, similarity search, Spark, or distributed training
