Senior Deep Learning Compiler Engineer - PyTorch
Munich, Bavaria, Germany · Full Time
Be the first to apply
- Experience
- 8+ yrs
- Salary
- PLN 292,500 – PLN 507,000 / year
- Openings
- 1
- Posted
- 4 hours ago
- Work mode
- In office
- Education
- Bachelor's or higher in Computer Science or related technical field (or equivalent experience)
- Resume
- Required to apply
Where you'll work
Sign in to tell us what does and doesn't work for you here — it sharpens every match we show you.
Job description
About the Role
Join NVIDIA at the cutting edge of AI compiler innovation and contribute to shaping the future of accelerated computing. The team is developing Thunder, a cutting-edge, source-to-source compiler designed to unlock exceptional performance for PyTorch models on NVIDIA GPUs. This opportunity involves collaborating with modern compiler frameworks, including PyTorch 2.0's TorchDynamo and TorchInductor, to create powerful open-source tools that benefit global AI researchers and developers.
Key Responsibilities
As a pivotal team member, you will lead development and optimization of core compiler technologies accelerating extensive deep learning workloads. Responsibilities include designing, implementing, and maintaining advanced compiler components, performing comprehensive performance analysis across thousands of GPUs, and collaborating closely with engineers who developed PyTorch for NVIDIA hardware. You will also engage with a wide ecosystem of experts working on nvFuser, TVM, XLA, CUDA, and other systems to translate research innovations into impactful open-source solutions.
Qualifications and Experience
- Bachelor's, Master's, or Ph.D. degree in Computer Science or related technical discipline (or equivalent experience).
- At least 8 years of relevant professional experience.
- Proficient in Python, with experience building intricate, robust software systems.
- Experience working with deep learning frameworks like PyTorch or JAX, including understanding model structure and performance bottlenecks.
- Strong background in compiler theory, including concepts like abstract syntax trees (ASTs), static single assignment (SSA) form, program analysis, and code generation.
- Excellent interpersonal communication and teamwork abilities, especially in distributed and open-source environments.
Preferred Skills
- Contributions to deep learning compiler projects such as TVM, MLIR, or IREE, or development within deep learning frameworks.
- In-depth knowledge of PyTorch internals, especially the compiler stack components like TorchDynamo and TorchInductor.
- Experience with JAX-style functional transformations applied to compiler designs.
- Familiarity with parallel programming, distributed computing systems, and high-performance CUDA programming.
- Demonstrated active involvement in open-source communities through code contributions, design input, or mentorship roles.
Additional Information
NVIDIA is a leader in AI, High-Performance Computing, and Visualization technologies. The company values innovation, diversity, and a supportive work environment. Compensation is competitive and commensurate with experience and location.
Compensation
Base salary depends on location and experience. For candidates based in Poland, the salary range is between 292,500 PLN and 507,000 PLN annually.
Minimum education
Bachelor's Degree