Data Infra Engineer - Petabyte-Scale Storage & Video Pipelines

Greylock Partners; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Greylock Partners is partnering with an early-stage AI company to build a durable data foundation for large-scale video and multimodal data. We seek a senior engineer who has owned petabyte-scale storage and data platforms in production, shaping architectures with long-term impact.

You'll design storage, processing pipelines, and observability, collaborating with founders to turn early product needs into scalable, robust data infrastructure for researchers and agents.

Qualifications

• Direct ownership of petabyte-scale storage or data platforms in production.
• Strong distributed-systems judgment around storage, throughput, reliability, and cost.
• Experience building data systems for demanding workloads like ML training or large-scale analytics.
• Ability to investigate failures across system boundaries and improve design.

Responsibilities

• Design and build storage and processing systems for large volumes of video and data.
• Develop reliable pipelines for high-throughput multimodal processing.
• Make deliberate choices about data layout, lifecycle, and cost as volume grows.
• Build infrastructure linking captured activity to research experiments and production workflows.
• Establish observability and operating practices for data quality and performance.
• Work with founders to shape durable data architecture.

Skills

Petabyte-scale storage

Distributed systems

Data platforms

ML data infrastructure

Job description

Greylock Partners is partnering with an early-stage AI company to build a durable data foundation for large-scale video and multimodal data. We seek a senior engineer who has owned petabyte-scale storage and data platforms in production, shaping architectures with long-term impact.

You'll design storage, processing pipelines, and observability, collaborating with founders to turn early product needs into scalable, robust data infrastructure for researchers and agents.