Staff Data Engineer — Lead Data Flywheel & Equity

LVT (LiveView Technologies); Seattle (WA); On-site; USD 172,000 - 221,000; Full time

Benefits offered by this job

Health benefits

401k match

Flexible PTO

Job summary

LVT is seeking a Staff Data Engineer to own the data flywheel end to end, from logs and sensor telemetry to labeled training data and benchmark sets. This is a senior IC role with technical-leadership responsibilities, partnering with AI/ML research and MLOps teams.

You will design and govern dataset schemas, storage strategies, versioning, and self-serve data access, enabling researchers to train and evaluate models efficiently while ensuring data safety and provenance.

Qualifications

• 8+ years building large-scale data pipelines and data-lake or lakehouse systems.
• Experience with multimodal data and video analytics.
• Strong Python and Spark data-engineering skills.
• Knowledge of dataset versioning, schema evolution, and governance.
• Experience integrating PyTorch/Lightning data loaders and Spark analytics.
• Bachelor's or Master's in CS, Engineering, or related field (or equivalent).

Responsibilities

• Own end-to-end data flywheel—from raw telemetry to labeled training data and evaluation sets.
• Build and maintain layered dataset pipelines and a single source of truth for training.
• Oversee labels, annotations, governance, and data quality throughout the lifecycle.
• Collaborate with AI/ML research and MLOps to ensure robust data platforms.
• Mentor ICs and set data-engineering standards.

Skills

Data pipelines

Python

Spark

Lakehouse architecture

Data governance

Education

Bachelor's or Master's in Computer Science or Engineering

Tools

PyTorch/Lightning

Job description

LVT is seeking a Staff Data Engineer to own the data flywheel end to end, from logs and sensor telemetry to labeled training data and benchmark sets. This is a senior IC role with technical-leadership responsibilities, partnering with AI/ML research and MLOps teams.

You will design and govern dataset schemas, storage strategies, versioning, and self-serve data access, enabling researchers to train and evaluate models efficiently while ensuring data safety and provenance.