Senior Compiler Optimization Engineer – LLVM

NVIDIA; Maharashtra; On-site; INR 3,000,000 - 5,000,000; Full time

Job summary

NVIDIA is seeking a Senior LLVM Compiler Engineer to help develop compilers for our world-leading GPUs in Maharashtra, India. You will contribute to a re-targetable GPU compiler backend and work on advanced optimization techniques across DL, graphics, HPC, and autonomous driving workloads.

You will collaborate with experts, tackle register allocation, instruction scheduling, loop optimizations, and program analysis, while strengthening our CUDA, OpenGL/Vulkan, and system software.

Qualifications

• B.S. or higher degree in Computer Science/Engineering and 8+ years in compiler design.
• Strong background in program analysis and high level optimizations.
• Excellent C and C++ programming skills and experience with LLVM/GCC.
• Good understanding of processor architecture and ability to work in a global team.
• Self-motivated, with strong communication skills and ability to influence others.

Responsibilities

• Be part of NVIDIA re-targetable GPU compiler backend team.
• Work on register allocation, instruction scheduling, synchronization, and loop optimizations.
• Develop innovative solutions to optimize workloads in Deep Learning, graphics, HPC.
• Solve unique problems in compiling for massively parallel GPU architectures.
• Influence and work on new GPU architectures.

Skills

C/C++ programming

Compiler design

LLVM

Register allocation

Instruction scheduling

Program analysis

High level optimizations

Communication skills

Self-motivated

Team collaboration

Education

B.S. or higher in Computer Science/Engineering

Masters or PhD preferred

Tools

LLVM

GCC

CUDA

OpenGL / Vulkan

Job description

We are looking to hire a Senior LLVM Compiler Engineer for an exciting role developing compilers for our world leading GPUs. We craft outstanding compilers that realize the potential of NVIDIA GPU hardware for growing range of computational workloads, ranging from next generate graphics, deep learning, scientific computing, and self-driving cars. Our compiler organization makes its mark on every GPU and SoC product that NVIDIA builds. Would you like to be part of this best-in-class organization and lead the charge?

What You'll Be Doing

• Be part of NVIDIA re-targetable GPU compiler backend team
• Work on challenging problems in register allocation, instruction scheduling, synchronization, loop optimizations etc.
• Develop innovative solutions to optimize workloads in Deep Learning, Ray Tracing, HPC etc. by working with industry experts in each field.
• Solve unique problems seen only in a compiling for massively parallel GPU architecture.
• Get a chance to influence and work on new GPU architectures.

What We Need To See

• B.S. or higher degree in Computer Science/Engineering
• Proficient in compiler design with 8+ years of experience
• Strong background in program analysis and high level optimizations
• Excellent C and C++ programming skills
• Hands on experience in compiler infrastructures like LLVM/GCC
• Good understanding of processor architecture
• Experience in a working in a global team
• Self-motivated to get things done
• Good communication skills with ability to influence others

Ways To Stand Out From The Crowd

• Masters or PhD preferred
• Demonstrated ability to pursue complex optimization problems
• Experience in Scheduling/Register Allocation
• Experience in developing CUDA, DirectX, OpenGL/Vulkan applications

NVIDIA is committed to developing a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.