Senior Staff Software Engineer — Agentic Workflows

NVIDIA•Us, CA

About The Position

Core Engineering Products in NVIDIA’s Platform Engineering organization supports software delivery, model publishing, and AI development. NVIDIA is hiring a Senior Staff Software Engineer to architect and develop AI automation that helps engineering teams address issues, retrieve services, and complete development workflows more efficiently. The role leads technical direction, architecture, implementation, and quality for production automation across these platforms. It is a hands-on software engineering position, with most time spent coding. The engineer will establish the long-term technical vision for AI automation and agentic workflows, translating recurring developer needs into reusable capabilities. They will own projects end to end, build agents and services that interpret requests and take approved actions, and automate onboarding, provisioning, publishing assistance, and troubleshooting. The role also focuses on reliable and safe execution through testing, evaluation, observability, retries, recovery, human approval, identity-aware authorization, credential protection, auditability, and data-access boundaries. The engineer will lead work across platform engineering, security, identity, and developer teams; establish shared interfaces and practices; mentor engineers; and measure results such as reduced manual effort, faster request completion, workflow success, developer adoption, and reliability.

Requirements

10+ years of software engineering experience building and operating production services, APIs, and distributed systems.
Bachelor’s or Master’s degree in Computer Science, Software Engineering, or a related technical field, or equivalent experience.
Demonstrated ownership of architecture and delivery for complex software projects, including defining technical direction with minimal mentorship.
Proficiency in Python, Go, or another backend language, with strong practices in testing, code review, CI/CD, and production debugging.
Experience building LLM applications or agentic workflows, including tool integration, retrieval, evaluation, and failure handling.
Ability to build secure integrations involving enterprise identity, permissions, credentials, and sensitive data.
A record of leading cross-team technical initiatives, communicating complex decisions clearly, and mentoring engineers.

Nice To Haves

Experience delivering AI automation with measurable reductions in support effort or developer wait time.
Familiarity with artifact repositories, model registries, experiment tracking, and software delivery workflows.
Experience with enterprise workflow connections such as ServiceNow and Jira, and platforms including Artifactory, Hugging Face, and CDN.
Experience crafting reusable agent tools and integrations using APIs, event-driven systems, or Model Context Protocol.

Responsibilities

Establish the long-term technical vision for AI automation and agentic workflows, turning recurring developer needs into reusable software capabilities.
Own software projects from architecture and implementation through deployment and ongoing improvement, making technical decisions that determine delivery success.
Build agents and services that interpret requests, retrieve relevant knowledge, diagnose issues, and complete approved actions across platform APIs and enterprise systems.
Automate onboarding, provisioning, publishing assistance, and troubleshooting.
Engineer reliable execution through evaluation, testing, observability, retries, recovery, and human approval where required.
Build identity-aware authorization, credential protection, auditability, and data-access boundaries into agent workflows.
Lead delivery across platform engineering, security, identity, and developer teams, establishing shared interfaces and engineering practices.
Mentor engineers and guide build and code reviews to improve software quality and delivery capability.
Measure outcomes including reduced manual effort, faster request completion, successful workflow execution, developer adoption, and reliability.