Senior Deep Learning Software Engineer, Inference and Model Optimization
Core
Develop and scale automated inference and deployment solutions for generative AI models (LLMs, diffusion models) to maximize efficiency.
Role type
Senior Deep Learning Software Engineer (Inference & Optimization)
Builds
Automated deployment solution (TRT Model Optimizer), high-performance inference kernels, and scalable software platforms.
Domain
Generative AI, Deep Learning, GPU Computing
Deliverable
production ML models
Required skills
Deep Learning fundamentals, Python, PyTorch, CUDA, GPU kernel development, Model optimization (pruning, quantization, sparsity), Distributed training (tensor/sequence parallelism), Performance profiling
Preferred skills
Contributions to PyTorch/JAX, Knowledge of GPU architecture and compilation stack, Familiarity with TensorRT, Experience with CUTLASS or Triton
Technologies
PyTorch, HuggingFace, CUDA, TRT-LLM, Triton, TorchDynamo, torch.export, torch.compile, TensorRT
Responsibilities
Develop high-performance optimization techniques for inference (sharding, attention kernels), Analyze and profile GPU kernel-level performance, Architect modular and scalable software platforms, Collaborate on performant kernel implementations, Deploy state-of-the-art generative AI models
Seniority
Senior, hands-on IC
