Tech Stack
Tag name is followed by "@" symbol and proficiency level value.
About proficiency levels:
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
AI @ 6
Agentic AI @ 6
Agentic Systems @ 6
CUDA @ 4
GPU @ 4
HPC
LLVM @ 6
Leadership @ 6
Mathematics @ 4
Performance Analysis @ 7
Performance Optimization @ 7
Python @ 1
Rust @ 1
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Details
We are looking for an outstanding Senior Compiler Engineer to help build the next generation of intelligent compiler technologies for NVIDIA's accelerated computing stack. The team works at the intersection of compilers, agentic systems, numerical correctness, and verification to create compiler systems that can reason about, generate, optimize, and validate code transformations across software and hardware boundaries.
In this role, you will help design and build advanced compiler and verification systems that improve developer productivity, system performance, and confidence in correctness. This includes developing compiler-oriented agents and workflows, building numerical validation and verification infrastructure, and co-designing with NVIDIA's compiler, runtime, library, and hardware teams to bring new capabilities into production tools and products.
Responsibilities
- Design, build, and optimize intelligent low-level compiler systems, including compiler-oriented agents, multi-step transformation workflows, tool use, memory, and orchestration.
- Develop systems that can reason about low-level IR, code generation, optimization passes, scheduling, numerical precision, and performance tradeoffs across GPU-centric workloads.
- Build the data, environments, verification, reward, and evaluation systems needed to continuously improve compiler quality, numerical correctness, robustness, and performance.
- Develop infrastructure for numerical validation, differential testing, regression detection, and correctness verification for generated or transformed code.
- Co-design compiler and verification workflows across NVIDIA's software stack, spanning models and inference, compiler pipelines, runtimes, libraries, kernels, and GPUs.
- Collaborate closely with internal compiler, CUDA, library, runtime, and hardware teams to integrate new compiler, agentic, and verification capabilities into NVIDIA products.
- Drive performance analysis and optimization for generated or transformed code, with a strong focus on correctness, scalability, and real-world deployment.
Requirements
- Bachelor's degree in Computer Science, Electrical Engineering, Applied Mathematics, or a related field, or equivalent experience; a master's degree or PhD is preferred.
- 5+ years of relevant industry or academic experience in compilers, AI systems, numerical computing, verification, high-performance computing, or related areas.
- Strong hands-on experience in compiler implementation, program analysis, code generation, optimization, low-level performance engineering, or correctness infrastructure.
- Experience in one or more of the following areas: agent systems, coding agents, compiler automation, numerical analysis, program verification, differential testing, or formal/semi-formal validation methods.
- Strong C/C++ and Python programming skills, with solid software engineering fundamentals; experience with Rust is a plus.
- Experience with GPU programming and performance optimization using CUDA or comparable accelerator platforms.
- Ability to work effectively across system boundaries, from model-driven workflows and compiler infrastructure to runtime and hardware-aware optimization.
Preferred Qualifications
- Track record of building high-impact compiler infrastructure, coding agents, verification frameworks, correctness tooling, or developer tools.
- Deep expertise in low-level compiler stacks such as LLVM, MLIR, PTX, Triton, or other IR/code generation frameworks.
- Experience building or optimizing AI-assisted compilation, numerical validation systems, equivalence checking, autotuning, or compiler verification workflows.
- Deep expertise in GPU systems, numerical behavior, and performance optimization, demonstrated through benchmark results, deployed systems, publications, or widely used software.
- Publications or open-source leadership in compilers, numerical computing, verification, formal methods, agentic AI, high-performance computing, AI systems, or programming systems.
Compensation and Benefits
- Base salary range: $152,000–$241,500 USD per year.
- Eligible for equity and benefits.
- NVIDIA offers a comprehensive benefits package.
Applications for this job will be accepted at least until September 21, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes. NVIDIA is committed to fostering an inclusive work environment and is an equal opportunity employer.