Senior Open Harness AI Engineer

NVIDIA Corporation; United States; A distancia; USD 184.000 - 288.000; Jornada completa

Descripción de la vacante

NVIDIA Corporation is seeking a Senior Software Engineer for Open Harness Engineering. You will build core libraries, evaluation systems, and reusable harness capabilities to improve AI agents’ safety, speed, and trust for developers.

This role spans open-source harnesses, sandboxed execution, and model-provider infrastructure with hands-on problem solving across multiple frameworks. You will diagnose failures via traces, logs, and artifacts, design reproducible experiments, and ship

Formación

• Bachelor’s or Master in Computer Science, Artificial Intelligence, Applied Math, or related field, or equivalent experience.
• 8+ years of hands-on software-engineering experience with production systems ownership and architecture leadership.
• Expert Python skills and proficiency in Rust, Go, C++, or TypeScript; able to debug across language boundaries.

Responsabilidades

• Define and evolve the technical approach for evaluating open agent harnesses across frameworks, models, and benchmarks.
• Build and operate scalable, reproducible evaluation systems spanning CI, sandboxed execution, and trace analysis.
• Diagnose agent failures from traces, logs, tool results, and benchmark artifacts across harness, environment, and model causes.
• Design controlled experiments and promotion criteria to distinguish genuine harness improvements from noise.
• Ship harness and runtime improvements, including upstream open-source contributions and documentation.
• Collaborate with Relay, Hermes, model, infrastructure, and research teams to turn goals into reusable capabilities.

Conocimientos

Python

Rust

C++

Go

TypeScript

Distributed systems

LLM agents

Evaluation infra

Educación

Bachelor’s or Master in CS/AI/Applied Math

Herramientas

OpenTelemetry

OpenInference

Descripción del empleo

NVIDIA Corporation is seeking a Senior Software Engineer for Open Harness Engineering. You will build core libraries, evaluation systems, and reusable harness capabilities to improve AI agents’ safety, speed, and trust for developers.

This role spans open-source harnesses, sandboxed execution, and model-provider infrastructure with hands-on problem solving across multiple frameworks. You will diagnose failures via traces, logs, and artifacts, design reproducible experiments, and ship