Nvidia Principal Engineer Agentic System Architecture 

United States, Texas 

496598990

NVIDIA is seeking a principal engineer to craft the architecture of systems that apply agents, agentic frameworks, and LLMs. This role focuses on designing, developing, and deploying scalable, high-performance systems that integrate autonomous agents built using existing and future frameworks. The ideal candidate will have a strong background in architecting frameworks and a shown ability to build robust platforms that address both production and research needs. As a Principal Engineer, you will lead the architectural design of end-to-end systems that incorporate multi-agent workflows, with the overarching goal of creating frameworks and reference architectures that allow enterprises to unlock PB+ scale of data. This AI query engine will quickly allow any developer and enterprise to implement applications that improve the efficiency and lives of employees. You'll work closely with teams from different functions to ensure the scalability, security, and performance optimization of this reference architecture while offering mentorship to engineers and encouraging innovation in the agentic systems space.

What you'll be doing:

• Architect and design large-scale systems of agentic frameworks (e.g., LangGraph, Llama Deploy, AutoGen, CrewAI) that form the base of an AI query engine.

• Develop system-level solutions that use autonomous agents with existing infrastructure, ensuring scalability, reliability, and composability.

• Evaluate and select appropriate foundational technologies for various use cases, ensuring optimal extensibility.

• Ensure system robustness by implementing security standard processes, including those for LLM safety and reliability and remain at the forefront of advancements in LLMs, agentic frameworks, and distributed system design.

• Provide technical leadership across projects and internal groups while mentoring engineers in system architecture design.

What we need to see:

• BS or MS in Computer Engineering, Computer Science, or a closely related quantitative field (or equivalent experience).

• 15+ years of experience in software engineering or system architecture with a strong focus on Python or C++ development.

• Proven track record of architecting production-level systems that use distributed computing

• Self-starter with excellent collaboration skills and a history of working with cross-discipline teams including research and production.

• Strong desire to work on the cutting edge of technology in a rapidly evolving environment while quickly learning and applying new technologies and libraries.

Ways to stand out from the crowd:

• PhD (or equivalent) in Computer Engineering, Computer Science, or a closely related field.

• Deep understanding of high-performance parallel computing, with experience in multi-threaded or multi-process environments.

• Familiarity with AI frameworks (e.g., PyTorch, TensorFlow) and NVIDIA technologies (e.g., CUDA, TensorRT, Triton) and experience developing for GPU platforms and understanding of GPU architectures.

• Proficiency in profiling Python code to optimize performance (e.g., asyncio vs threading vs multiprocessing).

• Demonstrated history of contributing and building open-source software projects that address real enterprise challenges at PB+ scale.