Senior Machine Learning Applications and Compiler Engineer, LPX

Overview

In this role you will advance NVIDIA’s inference and compiler stack by mapping neural workloads onto our platforms. You’ll work at the intersection of large-scale systems, compilers, and deep learning to optimize end-to-end inference performance. You will collaborate with hardware teams to shape future architectures and publish results at top venues. This is a hands-on position focused on practical, scalable techniques for real-world deployments.

Responsibilities

• Build and maintain high-performance runtime and compiler components for end-to-end inference optimization
• Define mappings of large-scale inference workloads onto NVIDIA systems
• Extend NVIDIA’s software ecosystem with libraries, tooling, and interfaces for cross-platform model deployment
• Benchmark, profile, and monitor performance to ensure efficient graph-to-hardware mappings
• Collaborate with hardware architects to feedback software observations and influence architectures
• Prototype and evaluate new compilation and runtime techniques for spatial processors
• Publish and present technical work at ML, compiler, and computer architecture venues

Key requirements

• MS or PhD in CS, ECE, or related field, or equivalent with 5+ years of experience
• Strong software engineering in systems programming (C/C++ and/or Rust)
• Hands-on compiler or runtime development experience (IR design, optimization passes, code generation)
• Experience with LLVM and/or MLIR (building passes, dialects, integrations)
• Familiarity with TensorFlow, PyTorch, and ONNX for graph formats
• Solid understanding of parallel and heterogeneous architectures (GPUs, spatial accelerators)
• Strong analytical and debugging skills using profiling and benchmarking tools
• Excellent communication and collaboration across hardware, systems, and software teams

• strong collaboration
• clear communication
• problem-solving mindset
• LLVM
• MLIR
• IR design

…