Staff ML Data Engineer

LiveView Technologies•Seattle, WA

14h

About The Position

LVT is seeking a Staff Data Engineer to own the data flywheel that supports its AI systems, including logs, sensor telemetry, labels and annotations, evaluation datasets, and benchmark sets. The role will build and maintain a shared stack that transforms raw data into standardized, versioned, governed datasets used by AI teams for training and evaluation. It is a senior individual-contributor and technical-leadership position; formal people management is not required. The role partners with AI/ML research, ML platform/MLOps, annotation, edge, and infrastructure teams. It owns the data-side contracts defining what models consume and emit, including dataset schemas, storage and partitioning approaches for multimodal data, versioning and migration strategies, and governance controls for sensitive video and sensor data. The position also establishes data-engineering standards and mentors individual contributors as the team grows. Candidates should bring substantial production data-engineering experience, ML training and evaluation data expertise, modern lakehouse architecture knowledge, multimodal data experience, Python and ML framework integration skills, and practical governance and provenance experience. A bachelor's or master's degree in a related field, or equivalent practical experience, is required.

Requirements

8+ years of experience building and operating large-scale production data pipelines and data-lake or lakehouse systems, including ingestion, ETL/ELT, partitioning, storage-format decisions, and reader/writer libraries.
Experience building data pipelines for model training and evaluation, including labeled data and evaluation or benchmark sets, and a working understanding of how data quality and versioning affect model results.
Strong experience with medallion-style layered data architectures and modern table or lake formats such as Iceberg, Delta, or Parquet, including schema evolution and dataset versioning.
Experience with large-scale multimodal data such as video, images, and sensor or telemetry data, including scalable storage and access patterns.
Hands-on experience with the data side of ML frameworks, including PyTorch/Lightning dataloaders and Spark, and strong Python knowledge.
Practical experience enforcing data governance in pipelines, including classification, access control, lineage and provenance, and retention, particularly for privacy-sensitive data.
Track record of setting data-engineering direction and developing engineers through technical leadership; formal management experience is not required.
Bachelor's or master's degree in Computer Science, Engineering, or a related field, or equivalent practical experience.
Must be authorized to work in the U.S.
Must pass a drug screening and background check upon employment.

Nice To Haves

Experience with streaming or near-real-time ingestion from edge or IoT sources into a data lake, using Kafka, Lambda, EMR, or similar technologies.
Experience with append-without-rewrite and hash-indexed dataset techniques on open table formats, and dataset or feature-versioning systems.
Generative-AI data experience, including fine-tuning and evaluation dataset curation for LLMs or VLMs.
Experience exposing datasets to AI agents through MCP-style query interfaces, with semantic schemas and plain-language documentation for retrieval.
Experience with computer-vision or video annotation tooling and workflows such as Encord or Labelbox.

Responsibilities

Own the end-to-end data flywheel that converts raw edge telemetry and video into labeled training data and frozen evaluation sets, and feeds model outputs into subsequent data cycles.
Build and own layered pipelines that register raw data, standardize it into a consistent schema, and produce curated datasets in a shared store.
Own the lifecycle, versioning, quality checks, and serving of labels and semantic annotations while partnering with annotation and data-operations teams on production and verification.
Own frozen, versioned validation and benchmark datasets and the review and scrubbing processes required before external sharing.
Manage schema and content versioning, including opt-in versions, append-without-rewrite changes, and controlled reader/writer migrations.
Build and maintain read/write libraries and integrations for PyTorch/Lightning dataloaders, record-level CRUDL APIs, and Spark/analytics access, enabling self-service for AI teams.
Enforce data governance in the data flywheel, including classification, scrubbing and anonymization, access controls, lineage, and provenance.
Set data-engineering standards for schemas, dataset contracts, and quality gates, and mentor individual contributors.