Senior DL Algorithms Engineer - Inference Performance
About the role
NVIDIA is seeking senior engineers to optimize deep learning workloads. This role involves working across the hardware and software stack to achieve peak performance.
Responsibilities
- Implement language and multimodal model inference as part of NVIDIA Inference Microservices (NIMs).
- Contribute new features, fix bugs, and deliver production code to TRT-LLM, NVIDIA's open-source inference serving library.
- Profile and analyze bottlenecks across the full inference stack to push the boundaries of inference performance.
- Benchmark state-of-the-art offerings in various DL models inference and perform competitive analysis for NVIDIA SW/HW stack.
- Collaborate with other SW/HW co-design teams to enable the creation of the next generation of AI-powered services.
Requirements
- PhD in CS, EE or CSEE or equivalent experience.
- 5+ years of experience.
- Strong background in deep learning and neural networks, in particular inference.
- Experience with performance profiling, analysis, and optimization, especially for GPU-based applications.
- Proficient in C++, PyTorch or equivalent frameworks.
- Deep understanding of computer architecture, and familiarity with the fundamentals of GPU architecture.
Qualifications
Proven experience with processor and system-level performance optimization.
Strong fundamentals in algorithms.
GPU programming experience (CUDA or OpenCL) is a plus.
Benefits
Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 184,000 USD - 287,500 USD for Level 4, and 224,000 USD - 356,500 USD for Level 5.
You will also be eligible for equity and benefits.
Pay
The base salary range is 184,000 USD - 287,500 USD for Level 4, and 224,000 USD - 356,500 USD for Level 5.
Schedule
Not specified.
Skills
Proven experience with processor and system-level performance optimization.
Strong fundamentals in algorithms.
GPU programming experience (CUDA or OpenCL) is a plus.