Staff Machine Learning Engineer, Data Curation & Dataset Strategy

Overview

As a data ingestion and provisioning lead, you define how driving data is scored, stored, and filtered to support model training at scale. You’ll shape the dataset strategy and partner with multiple teams to ensure training data is valuable and cost-efficient. You’ll translate strategy into production features, balancing quality and storage with ML-driven improvements. You’ll work closely with cross-functional teams to steer data policy and data provenance across the company. This is a hands-on, high-ownership role in a mission to scale Wayve’s autonomous driving data ecosystem.

Pay / Benefits

• competitive salary
• meaningful equity
• relocation support and visa sponsorship
• hybrid working model
• learning and development budgets
• comprehensive benefits

Responsibilities

• Build and own the ingestion-time scoring model and routing rules for data to hot, cold, archive, or delete
• Define data distributions in collaboration with training teams for each model
• Deliver storage-cost savings via heuristics and ML-driven filtering
• Partner with ingestion, annotation, retrieval, data quality, model training, AV/robotics, evaluation and storage teams on data policy
• Lead and contribute to data provisioning and data provenance features within Corpus
• Define dataset content strategy and metadata requirements to enable efficient data selection for training
• Drive data selection by model needs, prioritizing challenging or underrepresented scenarios
• Collaborate with Corpus Implementation Engineer to translate strategy into production systems
• Evolve from heuristics to ML-driven filtering for vehicle and edge data

Key requirements

• Strong autonomous-vehicle or robotics experience with large-scale driving or sensor data
• Hands-on ML engineer with a track record of scoring, selecting, or filtering data
• Experience operating at petabyte scale and related engineering challenges
• Proficient in Python and Apache Spark with large-scale cloud data experience
• Strong quantitative intuition guiding practical, measurable decisions
• Pragmatic, high-ownership, able to set direction across multiple teams without formal authority
• Experience in data selection, retention or filtering for model training, edge/vehicle-side filtering, computer vision, or high-volume IoT data; leading small cross-functional groups is a bonus

• cross-functional collaboration
• ownership and initiative
• pragmatism
• Python
• Apache Spark
• large-scale cloud data

…