PhD Large Language Model Engineer Co-op

What you'll do

• Conduct research on scalable training and inference of large language models using ML systems and HPC techniques.
• Develop and optimize distributed training frameworks and model parallelism strategies for large-scale AI workloads.
• Explore hardware-aware optimizations including quantization, sparsity-aware computation, and memory-efficient techniques.
• Implement and benchmark state-of-the-art ML system optimizations using high-performance computing techniques.
• Perform algorithm-hardware co-optimization to enhance the efficiency and performance of LLMs on modern hardware.
• Collaborate with research and engineering teams to publish findings in top-tier academic conferences.

What we're looking for

• Currently pursuing a PhD in Computer Science, Electrical Engineering, or a related field focusing on ML Systems, HPC, or AI Infrastructure.
• Strong background in machine learning, distributed systems, and parallel computing.
• Experience with deep learning frameworks such as PyTorch, TensorFlow, or JAX and large-scale model training.
• Proficiency in Python and C++ with experience in performance profiling and optimization.
• Knowledge of GPUs, ASICs, and distributed training paradigms like FSDP, ZeRO, DeepSpeed, or Megatron-LM.
• Familiarity with HPC techniques including MPI, Rcom/CUDA, RCCL/NCCL, and high-speed networking technologies.
• Prior research experience in scalable deep learning systems, large-scale LLM training, or AI acceleration.
• Experience with AI compiler optimizations such as Triton, XLA, or MLIR.