Deep Learning Software Engineer, TensorRT Performance

This is a fully remote position, open to applicants in California.

📋 Description

• Develop methodologies for performance benchmarking and analysis workflows for NVIDIA's inference ecosystem.

• Detect performance bottlenecks and identify optimization possibilities within TensorRT, TensorRT-EdgeLLM, and Torch-TensorRT.

• Contribute to features and code for NVIDIA and open-source inference frameworks.

• Create optimized model pipelines that involve quantization, scheduling, memory management, and distributed inference.

• Collaborate with various teams in generative AI, automotive, robotics, image understanding, and speech understanding domains.

• Enhance the performance of deep learning models across different architectures and NVIDIA accelerators.

⛳️ Requirements

• Bachelor's, Master's, PhD, or equivalent experience in relevant fields such as Computer Science, Computer Engineering, EECS, or AI.

• A minimum of 2 years of relevant experience in software development.

• Proficient in C++ programming.

• Proficient in Python programming.

• Strong software engineering capabilities.

• Familiarity with deep learning frameworks (e.g., PyTorch, JAX, TensorFlow, ONNX) and inference libraries (e.g., TensorRT, TensorRT-LLM, vLLM, SGLang, FlashInfer).

• Experience in performance analysis and optimization.

• Solid foundation and architectural understanding of GPUs (standout).

• In-depth knowledge of modern deep learning models and workloads (standout).

• Expertise in a deep learning programming domain-specific language such as CUDA, TileIR, CuTeDSL, cutlass, or Triton (standout).

• Previous contributions to major LLM inference frameworks or experience with graph compilers in deep learning inference (standout).

• Prior experience in optimizing performance for low-latency, resource-constrained systems or embedded AI pipelines (standout).

🏝️ Benefits

• Equity

• Benefits