Software Solution Architect, NVIS

Builds LLM-powered agents and automation tools for NVIDIA's internal delivery organization, translating operational needs into agentic workflows, RAG systems, and backend services. Works across software development, product, and field teams to ship reliable AI systems that reduce manual work and improve visibility. This is internal-facing rather than customer-deployed, but owns the full stack of agent development from composition through production.

As a Software Solution Architect, NVIS at NVIDIA, you will lead the transformation of AI infrastructure. Our NVIS team focuses on developing the next generation of NVIS Central, an agentic software platform with tools, services, and AI agents that automate, simplify, and speed up the work of our delivery organization. This role offers an outstanding chance to create and build LLM-powered agents that improve execution visibility, cut down manual tasks, and standardize workflows. These efforts allow NVIS to grow quickly and with high quality. Join us to bring up, validate, optimize, and upgrade large-scale AI Factory infrastructure for some of the world’s most advanced accelerated computing environments! What you'll be doing: Compose, build, and productionize agentic AI solutions, tools, and applications for the NVIS delivery organization. Develop LLM-based agents, skills, tool-calling workflows, orchestration logic, backend services, APIs, data pipelines, and automation features as part of NVIS Central. Translate field, delivery, operations, and product needs into clear technical builds, agent workflows, and working software. Develop agents that can reason across project data, knowledge bases, operational systems, logs, reports, and delivery workflows. Build workflows that help NVIS teams identify risks, summarize project status, automate repetitive tasks, improve readiness visibility, and simplify handoffs. Work with timely engineering, retrieval-augmented generation, context management, agent memory, function calling, evaluations, and guardrails to build reliable AI systems. Integrate LLMs and agents with internal systems, project data sources, knowledge repositories, reporting tools, and operational workflows. Collaborate closely with software developers, architects, product managers, DevOps/SRE, and NVIS field teams to successfully implement reliable and scalable solutions. Contribute to engineering guidelines, including code quality, testing, CI/CD, observability, documentation, security, and production support. What we need to see: B.Sc. degree or equivalent experience in Computer Science, Computer Engineering, or a related technical field. 2+ years of hands-on software development experience building production applications, platforms, automation tools, or AI-based systems. Strong programming experience with Python and modern backend development. Hands-on experience working with LLMs, agentic workflows, timely composition, tool/function calling, RAG, and AI application development. Experience crafting and implementing RESTful APIs, data services, workflow automation, and integrations with enterprise systems. Experience building reliable software around non-deterministic AI systems, including testing, evaluation, monitoring, and failure handling. Experience with Docker, Kubernetes, CI/CD, Git, observability, and cloud-native development practices. Background with SQL and NoSQL databases, data modeling, querying, indexing, and data integration. Excellent problem-solving skills, ownership attitude, and ability to operate in a fast paced, cross-functional environment. Ways to stand out from the crowd: Experience building agent platforms, copilots, multi-agent systems, tool-calling workflows, evaluation frameworks, or MCP-style integrations. Deep understanding of LLM application patterns such as context engineering, retrieval quality, timely/version management, agent planning, human-in-the-loop workflows, and AI safety guardrails. Experience with AI infrastructure, HPC clusters, NVIDIA DGX systems, SuperPOD, Spectrum-X, Ethernet, InfiniBand, Kubernetes, or SLURM. Experience in automating workflows related to field, delivery, operations, or professional services. Strong Linux, networking, security, SRE, or distributed systems background.