Deep Learning Compiler Verification Engineer
$116k - $190k • US, CA, Santa Clara • Full-time
Posted 22h ago
Remote Work Policy
On-site
Employment Type
Full-time
Categories
LLM Engineer
About the job
NVIDIA is seeking an engineer to develop and maintain compiler verification software and related infrastructure for accelerating deep learning workloads. This role involves solving critical problems in GPU computing and systems software, working alongside a diverse team. You will contribute to the next generation of compiler technologies, ensuring correctness across graph transformations, IR lowering, and GPU execution. If you are passionate about tackling complex challenges in AI and compiler development, we encourage you to apply.
Responsibilities
- Develop and maintain systems for verifying correctness in deep learning compilers, including graph transformations, IR lowering, and GPU execution.
- Analyze and validate compiler optimizations (e.g., graph rewrites, fusion passes, mixed-precision transformations) to ensure semantic and numerical preservation.
- Implement and extend test generation systems using deep learning solutions and analysis methods to explore model topologies, precision modes, and hardware targets.
- Contribute to improving the measurement of functional quality and performance for evolving models, compiler stacks, and hardware.
Requirements
- BS, MS, or PhD in Computer Science, Computer Engineering, Mathematics, or equivalent experience.
- 2+ years of software engineering experience, with a focus on compiler development, deep learning systems, or compiler verification.
- Proficiency in Python or C++.
- Experience with a major DL framework (e.g., PyTorch, JAX/XLA, TensorRT) involving model execution, graph representation, or runtime behavior.
- Strong problem-solving and debugging skills, with the ability to reason about model behavior and compiler-generated code.
Benefits
- Equity
- Benefits