Senior Engineer - AI Agents and Systems

What You Will Be Doing

• Local Inference Optimization: Optimize performance of local LLMs (Nemotron and others) on GeForce RTX hardware. Profile and optimize inference across Ollama, llama.cpp, and vLLM, minimizing latency and memory footprint using TensorRT and CUDA.
• Agent Runtime Engineering: Build and optimize agentic harnesses (NemoClaw, OpenClaw) to run natively and reliably on Windows. Implement the orchestration logic that lets multi-agent systems plan, act, and use tools efficiently on constrained consumer hardware.
• Sandboxing & Security: Implement policy-based privacy and security frameworks for autonomous agents, handling filesystem access, secure inference routing, and network egress within thorough sandboxed execution environments.
• Hardware/Software Integration: Work close to the metal, integrating agent and inference stacks with NVIDIA's driver and middleware layers to extract maximum performance from RTX GPUs.
• Cross-Team Collaboration: Partner with internal AI research teams, driver teams, and the open-source OpenClaw community to ensure our consumer hardware is the best possible platform for local agents.
• Code Quality: Write reliable, production-ready code, contribute to engineering best practices, and raise the technical bar through code review and design input.

What We Need To See

• Experience: 12+ years of relevant professional software engineering experience, with a track record of shipping performance-critical systems.
• Education: BS, MS, or PhD in Computer Science, Computer Engineering, or a related technical field (or equivalent experience).
• AI & GPU Infrastructure: Hands-on experience with LLM inference pipelines (Ollama, llama.cpp, vLLM), GPU-accelerated computing (CUDA, TensorRT), and running local models on consumer-grade hardware.
• Agentic Frameworks: Practical experience with modern agentic frameworks (e.g., OpenClaw, LangChain, AutoGPT) and a working understanding of how multi-agent systems plan, act, and use tools.
• Systems & OS Knowledge: Strong understanding of Windows OS internals, process isolation, sandboxing technologies, and system-level security.
• Programming Languages: Proficiency in C++ (performance-critical systems and OS integration), Python (AI and orchestration logic), and TypeScript (agent plugins and tooling).
• Communication: Ability to translate complex technical decisions into clear documentation and collaborate effectively across diverse engineering teams.

Ways To Stand Out From The Crowd

• Demonstrated open-source contributions to AI agent platforms or inference/orchestration tools (especially OpenClaw or llama.cpp).
• Deep knowledge of NVIDIA GeForce RTX architecture and its specific constraints and advantages for edge AI.
• Experience building virtualization, containerization, or sandboxing tools natively for Windows.
• Active technical community presence (blogs, talks, whitepapers) at the intersection of AI, security, and local compute.

About the Role

This is a deeply technical, code-first role. You will spend your days profiling inference pipelines, squeezing latency and memory out of local models, and hardening agent runtimes.

Benefits

NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most talented people on the planet working for us. As part of our team, you will have the opportunity to influence the future with your vision and expertise.

Pay

Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 224,000 USD - 356,500 USD for Level 5, and 272,000 USD - 431,250 USD for Level 6.

Schedule

You will also be eligible for equity and benefits.

Application Instructions

Applications for this job will be accepted at least until July 10, 2026. This posting is for an existing vacancy.