Data Systems Engineer — Low-Latency AI Telemetry

NVIDIA Corporation; Santa Clara (CA); On-site; USD 224,000 - 357,000; Full time

Job summary

NVIDIA Corporation in Santa Clara, CA is seeking a Software Engineer, Data Systems to build the store, data model, and query path for engineering performance data in GPU data centers, with billions of events per day and low latency.

You will join profiling data with context from hosts, schedulers, configuration, logs, and storage, ensuring traceability across jobs and machines. The role involves data contracts, performance tuning, and API development.

Qualifications

• Bachelor's degree in CS/CE or equivalent experience.
• 12+ years of software engineering delivering data pipelines and analytical services.
• Experience with large-scale production data and billions of events per day.
• Low-latency streaming pipelines with backpressure, replay, and deduplication.
• Judgment on event identity, ordering, and data transformations.
• Concrete performance investigation with baseline, measurements, and trade-offs.
• Data contracts: versioning and guiding consumers through changes.
• Debugging across components with tests and measurements.
• Clear communication about design choices and data constraints.
• Proficiency in C++, Python, and SQL.

Responsibilities

• Develop store, data model, and query path for engineering performance data at GPU-scale with low latency.
• Join profiling data with hosts, schedulers, config, logs, and storage for cross-job comparisons.
• Take ownership of data contracts, versioning, and smooth consumer transitions.
• Trade off data quality, latency, storage cost, and retention; verify improvements with measurements.
• Handle delayed, duplicated, incomplete, or inconsistent data and expose limitations.
• Develop the data API and generated views; implement access controls and stability under high demand.

Skills

C++

Python

SQL

Go

Time-series data

Education

Bachelor's degree in Computer Science, Computer Engineering, or related field

Tools

ClickHouse

Arrow

Parquet

Job description

NVIDIA Corporation in Santa Clara, CA is seeking a Software Engineer, Data Systems to build the store, data model, and query path for engineering performance data in GPU data centers, with billions of events per day and low latency.

You will join profiling data with context from hosts, schedulers, configuration, logs, and storage, ensuring traceability across jobs and machines. The role involves data contracts, performance tuning, and API development.