Senior LLM Agents Architect

What You'll Be Doing

As a Senior LLM Agents Architect at NVIDIA, you will develop innovative AI flows to improve hardware and software. You will work with a diverse team including hardware architects, verification engineers, GPU performance experts, and software developers to build and deploy agentic systems.

• Develop AI flows enhancing simulation analysis and GPU-kernel optimization.
• Facilitate workshops to translate expert insights into actionable tasks.
• Prototype and productize solutions integrating internal services and GPU capabilities.
• Implement evaluation frameworks with offline and online telemetry.
• Mentor teams with documentation, playbooks, and agent orchestration insights.

What We Need To See

This role requires over 7 years in applied ML/AI or large-scale systems, with a demonstrated track record in building agentic or LLM-powered applications in production. A Bachelor’s in Computer Science or Electrical Engineering is required, along with strong software engineering skills in Python and a systems language such as C++ or Rust. Experience with GPUs, CUDA, PyTorch or TensorFlow, and developing evaluation metrics is essential. Candidates must excel in collaboration, problem solving, and independent work.

Key skills/competency

• LLM
• AI
• Agents
• Architecture
• Simulation
• Optimization
• Python
• GPU
• Prototyping
• Collaboration